Full Proceedings
Full Proceedings
com
Editors
Prof. (Dr.) Harish Kumar
Prof. (Dr.) V.K. Sharma
ISBN: 978-93-82880-20-2
No part of this publication may be reproduced or transmitted in any form by any means, electronic or
mechanical, including photocopy, recording, or any information storage and retrieval system, without
permission in writing from the copyright owners.
DISCLAIMER
The authors are solely responsible for the contents of the papers compiled in this volume.
The publishers or editors do not take any responsibility for the same in any manner. Errors,
if any, are purely unintentional and readers are requested to communicate such errors to the editors or
publishers to avoid discrepancies in future.
Published by
EXCEL INDIA PUBLISHERS
91 A, Ground Floor
Pratik Market, Munirka, New Delhi–110067
Tel: +91-11-2671 1755/ 2755/ 3755/ 5755 ● Fax: +91-11-2671 6755
E-mail: publishing@groupexcelindia.com
Web: www.groupexcelindia.com
Typeset by
Excel Publishing Services, New Delhi–110067
E-mail: prepress@groupexcelindia.com
Printed by
Excel Printing Universe, New Delhi–110067
E-mail: printing@groupexcelindia.com
Sponsored by
Co-sponsored by
Technically Sponsored by
Message
Welcome to TSPC 13
It is a great pleasure to welcome all of you to annully conference TSPC-13 at Bhagwant
Institute of Technology, Muzaffarnagar India. which is one of the countries recently
drawing much attention in the world arena of the most advanced ICT convergence
infrastructure.
Conference TSPC-13 is financial sponsored by IEEE-UP section & INSA, Govt of
India, CST & technical sponsored by BSNL, DST, Agilent, Texas Instruments India and
also collaboration with GBTU Lucknow & MTU Noida. TSPC,13 has been providing the
scientists, researchers and engineers policy makers, network planners, and service
providers in telecommunications from all over the world a scientific forum of disscussions
to exchange ideas and the research results as to develope new scientific discoveries in the relevant area of interest &
extensive exchange of information will be provided on newly emerging systems, standards, regulations, services,
and variety of applications in the area of communications & signal processing field, due to rapid penetration of
various smart devices such as smart phones, the network traffic is dramatically increasing and the network market is
changing recently. The network must be smarter in order to keep up with the change of network environment and
the advent of future smart services. To deal with these technology and market trends, we chose the phrase “Smart
Service Innovation through Mobile”.
Interactivity is the theme of TSPC-13. Under this theme, TSPC-13 consists of a plenary session with five
keynote speeches by remarkable world level experts, 03 tutorial sessions with interesting subjects, and 75 technical
sessions including 02 poster sessions. The TSPC carefully selected a limited number of 140 papers from 16 state to
be presented in the technical sessions. We hope these technical sessions will help you obtain better insights into the
latest trends and directions for the competitive broadcasting and telecommunications network markets. We believe
that TSPC-13 will provide an abundance of intelligent debate and information exchange, and a look at what is new
in the emerging field and technology.
I would like to express my deep appreciation and gratitude to all the program committee members, authors, the
reviewers, and advisory committee and arrangement & registration committee members for their terrific work in
collecting and editing the papers witin time on behalf of the Organizing and Technical Program Committees, Last
but not least, I would like to thanks all the plenary speakers, and fellow researchers that has been willing to share
their research knowledge and experience to all of us in this prestigious conference.
I hope the TSPC 13 will bring to you an unforgetable beautiful experience of enjoyment of your time in BIT.
Organizing Committee
Prof. Harish Kumar Conference Chair
Prof. H.P. Singh Principal
Dr. Ajay Gupta Asst. Director
Mr. D.V. Singh Dy Registrar
C.A. Dushayant Kumar Account & Finance Officer
ix
Committees
Volunteers
Vivek Garg Shalni Gupta
Diksha Mehlotra Vandana
Meghna Khanna Richa
Pranjal Rastogi Rida Batul
Parnav Chaudhary Shakti Singh
Ruchi Priya Subhash Chandra
Supriya Bharti Bhanu Singh Rathore
Ratna Priya Kanchan Diksha Verma
Chaturanan Pathak Arvind Verma
Kundan Kumar Nitin Chauhan
Arun Bhardwaj Mudita Sharma
Shiv Pratap Tanya Agarwal
Vidit Mishra Shaurabh Singh Chauhan
Shivani Rajput
Technical Committee
Prof. Dharmendra Singh IIT Roorkee
Prof. D.R. Bhaskar Jamia Millia University, Delhi
Prof. M.N. Doja Jamia Millia University, Delhi
Dr. Rajeev Kumar Kanth University of Turku, Finland
Prof. Akshay Kumar Rathor National Uinversity Singapore
Dr Manoj Datta GIFU University, Japan
Prof. A.P. Mittal NSIT Delhi
Prof. Avdhesh Sharma IIT Jodhapur
Prof. Dilip D. Choudhary Pune University
Prof. Sarvjeet Singh NIT Jalandhar
Prof. B.K. Koushik NIT Kurukhetra
Dr. Vishal Verma Delhi Tech. University
Prof. M.C. Govil Government Engineering College Ajmer India
Prof. Y.K. Vijay Rajasthan University Jaipur
Prof. P.K. Pauras Agra University
Dr. Dhananjay Singh National Institute for mathematical Sceince, Korea
Prof. Deepashri Pune University
Mr. Vikas Chola Taxas Instrument India
Prof Kriti Matur DAU Indor
Dr. Kuldeep Sharma BARC Mumbai
Prof. D.K.P. Singh Accurate Institute of Technology, Gr. Noida
Prof. J.K. Dwivedi HBTI Kanpur
Prof. Monaj Shukla HBTI Kanpur
Prof. Saurabh Raina JP University, Noida
Prof. K.M. Soni Amity University Noida
Prof. GK Singh, Galgotias University, India
Prof. Prasant Johri Galgotias University, India
Prof. K.K. Paliwal, NC College Panipat, India
Prof. Rajeev Tiwari HCST Mathura
Dr. Madur Deo Upadhayay IIT Delhi
Dr. Rabins Porwal ITS Ghaziabad
Prof. Ravi Kumar KVG Engineering College Manglore
Dr. Rajeev Singh Bhagwant University, Ajmer
x
TUTORIAL TRACK
Intelligent System Applications in Solving
Engineering Problems
Abstract—Due to increased complexity of the systems, Electricity Board as Assistant Engineer from 1988 to
cost reduction and detail modeling of the systems, the 1996, with Rookree University (Now IIT Rookree) as
requirements of optimization have been increased. The Assistant Professor from 1996 to 2001 and with Asian
conventional methods, which guarantee to provide the Institute of Technology, Bangkok, Thailand as Assistant
optimal solution, fail to solve many practical problems due
to intensive requirements of these methods. The most
Professor from 2001 to 2002. Dr Singh received several
familiar conventional optimization techniques fall in two awards including Young Engineer Award 2000 of
categories viz. calculus based method and enumerative Indian National Academy of Engineering, Khosla
schemes. Though well developed, these techniques possess Research Award of IIT Roorkee, and Young Engineer
significant drawbacks. Calculus based optimization Award of CBIP New Delhi (India), 1996. Prof Singh is
generally relies on continuity assumptions and existence of receipt of Humboldt Fellowship of Germany (2005,
derivatives. Enumerative techniques rely on special 2007) and Otto-monsted Fellowship of Denmark
convergence properties and auxiliary function evaluation. (2009-10).
Moreover, these optimizations are generally single path His research interests include power system
search and stuck with the local optima.
Intelligent based optimization methods such as
restructuring, FACTS, power system optimization &
genetic algorithm (GA), particle warm optimization control, power quality, wind power, etc. Prof Singh is a
(PSO), bacteria foraging, ant colony, neural networks etc Fellow of Institution of Electronics and
are multi-path search and provide solution near to global Telecommunication Engineers (IETE) India, a Senior
optima. This presentation briefly covers some of the Member of IEEE, USA, a Fellow of the Institution of
important techniques of optimization along with scope and Engineering & Technology (UK) and a Fellow of the
future challenges. Institution of Engineers (India).
Prof Singh has published more than 330 papers in
ABOUT SPEAKER International/national journals/conferences. He has also
Prof. S.N. Singh obtained his M. written two books one on Electric Power Generation,
Tech. and Ph. D. in Electrical Transmission and Distribution and second is Basic
Engineering from Indian Institute of Electrical Engineering, published by PHI, India.
Technology Kanpur, in 1989 and Prof Singh is Chairman, IEEE UP Section and
1995, respectively. Presently, he is a Chairman, IEEE PES/IAS joint Chapter of IEEE UP
Professor in the Department of Section. He is member, Board of Governors, IIT
Electrical Engineering, Indian Kanpur, MMMEC Gorakhpur and KNIT Sultanpur. He
Institute of Technology Kanpur, India. is also senate member of MNNIT Allahabad, MANIT
Before joining IIT Kanpur as Bhopal and IIT Mandi.
Associate Professor, Dr. Singh worked with UP State
Evaluating Sustainability, Environmental
Assessment and Toxic Emissions during
Manufacturing Process of RFID based Systems
Harish Kumar
Professor & Dean Academics,
Bhagwant Group of Institutions, Muzaffarnagar, India
Abstract—The present state of the art research in the environmental assessment in fabricating processes of the
direction of embedded systems demonstrate that analysis RFID based systems. We have chosen a general life cycle
of life-cycle, sustainability and environmental assessment assessment approach which involves the collection and
have not been a core focus for researchers. To maximize a evaluation of quantitative data on the inputs and outputs
researcher’s contribution in formulating environmentally of materials and energy associated with the RFID based
friendly products, devising green manufacturing processes systems. Based on the developed generic models, we have
and services, there is a strong need to enhance life-cycle obtained the results in terms of environmental emissions
awareness and sustainability understandings among for a production of paper substrate printed RFID
embedded systems researchers, so that the next generation antennas. We also make an attempt to raise several
of engineers will be able to realize the goal of a sustainable sustainability issues and quantify the toxic emissions
life-cycle. In this work an attempt has been made to during the manufacturing process.
investigate and evaluate the life-cycle management and
Mobile Communication Network Architecture
Dr. G.C. Manna and S. Chandran
Sr GM BSNL
BSNL, India
Abstract—A Communication Network consists of information among equipments before sending them
three building blocks: (1) Access Network- A sender alert messages. Further, Communication service is now
(and a receiver) network (2) Core Network-A complex transforming into a new business through deployment of
switching matrix to connect the sender to receiver and cognitive radio technology and spectrum bidding
(3) Transmission Network-the interconnection among concept.
several switches from sender end switch to receiver end The tutorial discusses the basics of radio access
switch. The success story of mobile communication is technologies deployed by mobile network operators viz.
based on revolution in VLSI technology together with GSM, CDMA and WiMAX and the user network access
embedded software support which has made all the equipment i.e. mobile handset and data card. To
equipments intelligent and service aggressive. understand connectivity between the end users, the
The information hungry world has lead telecom functions of different core network equipments , their
equipment manufacturers to develop fabulous service interconnection and signalling information exchange is
delivery platform to meet various type of customer explained step by step . A glimpse of future generation
needs. Thus, to provide instant services to the users, cognitive radio network is of interest to researchers to
Communication network is now grossly dependent upon formulate their thesis work.
efficient signalling system for quick exchange of
Contents
Messages vi
Committees viii
Tutorial Track xi
19. Analysis and Design of a Microstrip Patch Antenna with Circular Polarization
Ravindra Prakash Gupta, Yogesh Mehta, Surbhi Sharma and Virendra Kumar Sharma 66
TRACK II EMERGING TECHNOLOGY TRENDS
20. Impacts & Compensation of Distributed Generation using Smart Grid
Technology
Manoj Kumar Nigam and A. Krishna Nag 71
21. Precariousness in Technology Via Web Crime and Statically Crimes in the World
Ajay B. Gadicha 75
22. Performance Improvement of MOSFET with High-K Dielectric Materials
S.L. Tripathi, Priti Gupta and R.A. Mishra 79
23. A Review Paper on Single Electron Transistor and its Applications
Nitin Shera and Anil Kumar 82
24. Protection of Embedded Software from Side Channel Attacks using Virtual
Secure Circuit
H.K. Yasar Tharik and N. Neelima 85
25. On Capturing Exceptions in Workflow Models for Flexible Processes
Shivani Chauhan and Punit Agarwal 90
26. Analysis of Characteristics of Magnetic Resonant Coupling in Wireless Power
Transfer
Divya Vaishnav and Ritesh Diwan 96
27. Stepped Impedance Micro-strip Line Low Pass Filter for Microwave L-Band
Amit Kumar, Pradeep Rana, Subodh Kumar Tripathi and Manish Singh 100
28. A Simple Design of a Narrow Band TM-mode Filter using 1D Dielectric Photonic
Crystal
Deepti Jain, K.D. Jindal, Kh. S. Singh, Vipin Kumar and Sanjeev Sharma 103
29. Nanorobotics
K.B. Mohd. Umar Ansari and Goldy Sharma 106
30. Digital Image Watermarking
Ravindra Kumar 110
31. Wireless Sensor Network for Landslide Detection
Manish Kumar, Mayank Tripathi, Mohit Kumar and Anand Tiwari 113
32. A Tutorial Overview on the DGS to Design Microwave Devices
Vikash Kumar and Sanjay Gurjar 119
33. Study of Improved Technique of Bridging Fault Simulation
Diksha Verma 123
34. Survey on 4G Technology and its Applications
Khaleel Ahmad, Ankur Bhardwaj and Ashutosh Kumar 126
35. A Study of Compact Muon Solenoid
Avinash Kumar 130
36. Fabrication & Characterization of Graphite/ Epoxy Composites
Meenakshi Arya, Harish Kumar, M.G.H. Zaidi and Alkesh Chauhan 136
TRACK III SIGNAL PROCESSING AND PATTERN RECOGNITION
37. Signal Generation by Direct Digital Synthesis for Testing Fast Response System
for Eddy Current Flowmeter
Poornapushpakala S., Gomathy C., Sylvia J.I. and Babu B. 141
TRACK I
ANTENNA DESIGN FOR COGNITIVE RADIO APPLICATIONS
ISBN: 978-93-82880-20-2 3
Proceedings of National Conference on Trends in Signal Processing & Communication (TSPC’13), 12th–14th April 2013
Abstract—The purpose of this paper is two-fold. initiated antenna research in various directions, one of
Firstly, we have attempted to design a planar Sierpinski which is by using fractal shaped antenna elements.
fractal antenna in stacked configuration for multiband Fractal geometries have two common properties,
application. The stacked configuration of Sierpinski
fractal patch and Sierpinski grid areemployed to improve space-filling and self-similarity. The self-similarity
the multiband characteristics. The operating frequencies properties of certain fractals result in a multi-band
obtained are at 3.3 GHz, 5 GHz, 5.74 GHz and 5.9 GHz behavior [3-6], while the spacefilling properties of
which covers the bands useful for HIPERLAN2 fractals make possible to reduce antenna size [7].
frequencies and for implementation in futuristic WiFi Fractal objects have self-similar shapes, which mean
enabled devices and PCI Cards for mobile internet. The
Simulated results show that the operating frequencies that some of their parts have the same shape as the
obtained are spread over a wide range of frequency band whole object but at a different scale. In this paper the
compared to the simple Sierpinski fractal patch antenna. Sierpinski patch antenna has been considered for the
Secondly this paper aims at evaluating the sustainability investigation.
and life cycle management of proposed antenna. The Sierpinski gasket is named after the Polish
Keywords: Sierpinski fractal antenna, Sierpinski grid mathematician Sierpinski who described some of the
patch, multiband antenna, stacked configuration, main properties of this fractal shape in 1916 [8]. The
Sustainability Analysis, Life Cycle Assessment original gasket is constructed by subtracting a central
inverted triangle from a main triangle shape as shown
I. INTRODUCTION
in Fig. 1. After the subtraction, three equal triangles
The rapid progress in telecommunication remain on the structure, each one being half of the size
technology deals with a great variety of communication of the original one. This is the first iteration of the
systems like cellular, global positioning, and satellite Sierpinski fractal generation. One can iterate the same
communication systems. Each of these systems subtraction procedure on the remaining triangles and if
operates at several frequency bands employing a the iteration is carried out an infinite number of times,
number of small to large antennas depending upon the the ideal fractal Sierpinski gasket is obtained. In such
choices of applications. Printed antenna offers an an ideal structure, each one of its three main parts is
excellent performance in a wide variety of wireless exactly equal to the whole object, but scaled by a factor
communication applications. The recent trends
of two and so is each of the three gaskets that compose
establish evidence that production of antennas for
any of those parts. Due to this particular similarity
different applications are growing rapidly. Moreover,
due to low weight, low profile and low cost production property, shared with many other fractal shapes, it is
of patch and fractal structures, these antennas are said that the Sierpinski gasket is a self-similar [9-10].
becoming more popular in wireless and The following sections have been included in this
telecommunication industries. Various multiband paper. The next two sections describe the antenna
designs employing parasitic patches or shorting pins design criteria and the results obtained respectively.
have been proposed to date [1-2]. However, these Section IV deals with sustainability model and explores
techniques usually lead to an increase in antenna size or environmental impacts assessment. Section V would
manufacturing cost. This has give a conclusion of our work.
ISBN: 978-93-82880-20-2 3
Proceedings of National Conference on Trends in Signal Processing & Communication (TSPC’13) 12th–14th April 2013
contribution shows that the impact on emissions to [6] Tripti Luintel and Parveen F. Wahid, “Modified Sierpinski Fractal
Antenna”, IEEE/ACES International Conference on Wireless
fresh water is larger than emissions to air and sea
Communications and Applied Computational Electromagnetics, 3-7
water. April 2005, 578-581
[7] B. N. Biswas, Rowdra Ghatak, Rabindra K. Mishra, and Dipak R.
V. CONCLUSION Poddar, “Characterization of a Self-complementary Sierpinski
Gasket Microstrip Antenna”, PIERS ONLINE, vol. 2, no. 6, pp.
In this paper, we have attempted to give an 698-701, 2006.
overview of designing a fractal patch antenna in [8] R. K. Kanth, W. Ahmad, S. Shakya, P, Liljeberg, L.-R. Zheng,
HTenhunen “Autonomous use of fractal in structure low cost,
stacked configuration as well as environmental impacts multiband and compact navigation antenna in proceeding of MMS,
during its manufacturing process. The proposed aug.2010, Cyprus pp 135-138
antenna is designed in which a second stage of the [9] H. Kumar, Rajeev Kumar Kanth, P. Liljeberg, and H
Sierpinski gasket (1st iteration) is used as a the lower Tenhunen,“Metamaterial based slotted patch antenna,” in
Proceedings of the IEEE 10th International Conference on
patch and fed with coaxial probe, where as the Telecommunication in modern satellite Cable and Broadcasting
Sierpinski grid has been used as the upper parasitic Services, vol. 2, pp. 536–539, 2011
patch. This antennas has four operating frequency 3.3 [10] C. Puente, “Fractal antennas,” Ph.D. dissertation, Dept. Signal
GHz, 5 GHz, 5.75 GHz and 5.9 GHz. The major Theory Communicat., Universitat Polit´ecnica de Catalunya, June
1997.
applications of these antennas fall within area of [11] X. Wang, G. Gaustad, Golisano Institute of Sustainability, RIT,
HIPERLAN2, WiFi enabled devices and PCI Cards for Prioritizing Material Recovery for End-of-Life Printed Circuit
mobile internet. We have also explored the Boards,Proceedings of 2011 International Symposium on
environmental impacts in manufacturing process of this Sustainable Systems and Technology (ISSST)2011
[12] Maruschke J., Rosemann B., Measuring environmental performance
sierpinski type multiband printed microstrip antenna. in the early phase of product design using life cycle assessment,
Environmentally Conscious Design and Inverse Manufacturing,
ACKNOWLEDGEMENT 2005. pp 248-249
[13] V. Pekkanen, M. Mntysalo and P. Mansikkamki, Design
The authors acknowledge iPack Vinn Excellence considerations for inkjet printed electronic interconnections and
Centre, Royal Institute of Technology (KTH) for packaging, Proceedings of the Imaps 40th International Symposium
on Microelectronics. California (USA) (2007), pp. 1076-1083
supporting the necessary tools and equipments to carry [14] H. Griese, L. Stobbe, A. Middendorf and H. Reichl, Environmental
out these investigations. We would like to thank Turku compatibility of electronics-a key towards local and global
Centre for Computer Science for assisting the financial sustainable development, Proceedings of 2004 international IEEE
support. conference on the Asian green electronics (2004), pp. 158-163.
[15] Chelsea Bailey, Callie babbitt, and Gabrielle Gaustad, Rochester
Institute of Technology, Tracking the Material, Energy, and value
flow for end-of-life lithium ion Batteries in the USProceedings of
2011 International Symposium on Sustainable Systems and
Technology (ISSST)2011
[16] Rajeev Kumar Kanth, Pasi Liljeberg, Hannu Tenhunen, Wan
Qiansu, Yasar Amin, Botao Shao, Qiang Chen, Lirong Zheng,
Harish Kumar, Evaluating Sustainability, Environmental
Assessment and Toxic Emissions during Manufacturing Process of
RFID Based Systems. In: Jinjun Chen (Ed.), 2011 IEEE Ninth
International Conference on Dependable, Autonomic and Secure
Computing (DASC),, 1066-1071, DOI: 10.1109/DASC.2011.175
IEEE Computer Society, 2012.
[17] Rajeev Kumar Kanth, Pasi Liljeberg, Yasar Amin, Qiang Chen,
Fig. 6: Environmental Emissions during Manufacturing
Lirong Zheng, Hannu Tenhunen, Comparative End-of-Life Study of
of Sierpinski Multiband Antenna
Polymer and Paper Based Radio Frequency Devices. International
Journal of Environmental Protection 2(8), PP: 23–27, 2012. Online
REFERENCES ISSS: 2224-7777, Print ISSN:2226-6437
[18] Rajeev Kumar Kanth, Qiansu Wan, Harish Kuamr, Pasi Liljeberg,
[1] J.R. James and P. S. Hall, Handbook of Microstrip Antennas. Qiang Chen, Lirong Zheng, Hannu Tenhunen, Evaluating
London, U.K.: Peter Peregrinus, vol. 1, 1989. Sustainability, Environment Assessment and Toxic Emissions in
[2] Kumar H, Upadhayay MD, Rai VK, Varsheney Leena, Kanth RK, Life Cycle Stages of Printed Antenna. Elsevier, Procedia
“Multiband planar mircostrip antenna,” proceeding of the IEEE Engineering 30(1), PP: 508-513, 2012.
International conference on antenna & propagation in wireless [19] Rajeev Kumar Kanth, Pasi Liljeberg, Hannu Tenhunen, Qiang
communication, Sept. 16-20,2011, vol. 2, 978-1-4577-0048- Chen, Lirong Zheng, Harish Kumar, Comparative Toxic Emission
4/11//$26.00 ©2011 IEEE, pp 1360-1363 Analysis in Production Process of Polymer and Paper Based RFID
[3] S.R. Best, “On the Significance of Self-Similar Fractal Geometry in Tags. In: M. Caciotta, Zbigniew Leonowicz (Eds.), 11th
Determining the Multiband Behavior of the Sierpinski Gasket International Conference on Environment and Electrical
Antenna,” IEEE Antennas and Wireless Propagation Letters, vol.1, Engineering, PP: 184–187, DOI: 10.1109/ EEEIC.2012.6221570
no. 1, pp.22-25, 2002. IEEE, 2012.
[4] C. Puente, J. Ponmeu, R. Pous and A. Cardama, “On the behavior of [20] Rajeev Kumar Kanth, Pasi Liljeberg, Hannu Tenhunen, Qiang
the Sierpinski Multiband Antenna,” IEEE Transactions on Antennas Chen, Lirong Zheng, Haris Kumar, Study on Glass Epoxy Based
and Propagation, vol. 46, no. 4, pp.517-524, April 1998. Low Cost and Compact Tip-Truncated Triangular Printed Antenna .
[5] Junho Yeo and Raj Mittra, “Design of Conformal Multiband International Journal of Antennas and Propagation 2012
Antennas Based on Fractal Concepts”, Microwave and Optical (2012)(184537), PP: 1–8, 2012. Hindawi Publication,
Technology Letters, vol. 36, No. 5, 333-338, 2003. DOI:10.1155/2012/184537.
B
Band R nfiguraable Antenn
Recon A na
Satish Guupta1, Mithileesh Kumar1 and Upadhayyay M.D.2
1
U, Kota, 2Delhii, India
RTU,
Abstract—This papeer present thee idea of a simple • Carrier grade soluutions for hot zone
frequency reconfigurabble antenna that operatees at deploymeents
multiband 5.01GHz, 7.2G
GHz and wide band from 4.33 GHz
to 7.1 GHHz in this papper. By switch hing the diodde for • Indoor annd outdoor environments
microstrip elements on n the antenn na, the operrating 4.5 GHz for Body Area Network as describe in
frequency will shift witth fixed radiaation patterns.. The IEEE
E 802.15 TG66.
simulated return loss and
a surface current
c resultts are 6.5 & 6.7 GHz
G for Low
w power wireeless sensor
nd shown for both
provided an b cases. netw
work.
nt; formatting; style; styling; insert
Keywords: componen
(key words) II. ANTEN
NNA STRUCTU
URE AND DESIG
GN
I. INTRODUCTION
N The antenna structure andd dimension parameters
are shown
s in Fig.. 1. A 0.3 mm m gap is placced between
Recenntly, reconfiguurable antennaa has gained a lot the element
e and U shape parasiitic element; 0.50 mmx0.5
of attention in the anteenna design for f multi-funcction- mm microstrip connecting
c eelements which can be
oriented wireless
w commmunication appplications succh as regarrded as ideal switches, are placed on the t gap for
Software Defined
D Radio (SDR), Muultiple-in-Mulltiple- ideall ‘ON’ state and
a absence oof strip is useed for Ideal
out (MIM MO) and RFID tag etc.. Frequuency OFF F state. The transmission line model, is used to
reconfiguraability is one of the reconffigurable funcctions desiggn the propossed antenna att 5 GHz for thist purpose
(polarizatioon, radiation pattern, and operation
o bannd) in radiaating element strips of widdth 3.06mm, length 24.5
the reconffigurable anteenna design. At present, some mm and 29 mm with separation transmisssion line of
frequency reconfigurablle antennas haave been discuussed lengtth 7.335 mm m are designedd on FR4 (Er = 4.4) of
in earlier literatures
l [2]--[4]. Multi baand antenna caan be 1.6mmm thickness substrate.
s Thee third elemen nt has length
used for multiple
m commmunications siimultaneouslyy, but of 155.5mm with separation
s linne 7.0375mm. CPW feed
drawback of channel interferencee [1]. Frequuency hosen for widee band application. CPW ground
is ch g is of
reconfiguraability can bee achieved by varying the leength 5 mm m width and 10.75mm lenggth, CPW gap p is 1.3 mm
of radiatinng element. Foor this purposse some switcching and width
w of feed line is 4.75mmm. When all strips come
elements are used likke diode, traansistor. Diodde is into picture propposed antennaa works for wide band
preferred over
o the transsistor because it is two termminal appliication. Whenn third strip is cut off, anten
nna supports
device andd easy to biias. In this paper, p the deesign, two bands.
b
simulation of a simple s multtiband frequuency
reconfiguraable antennaa with microostrip elemennts is
discussed. It is found thatt the antennna can operaate at
multiband from 5.01GH Hz, 7.2GHz annd wide band from
4.3 GHz too 7.1 GHz annd the operatinng frequency band
can be shhifted when the elementt is connectt and
disconnectt using switchh on antenna. This antennaa can
be used in various appliccations like
5.2GHHz Wireless infrastructure:
i : WLAN, HL LAN,
DBS, broaddband, cellulaar base stationns
• Wi-Fi
W and picocell network
• Puublic wirelesss hotspots (ee.g. sports veenues,
hootels, airports,, etc.)
• 8002.11 a, b, g, n platforms
Fig. 1(aa): The Schematicc of the Antenna
• MIMO,
M widebaand and multi--band antennaa
ISBN: 978-9
93-82880-20-2 7
Proceedings of National Conference on Trends in Signal Processing & Communication (TSPC’13), 12th–14th April 2013
Fig. 4(a): Simulation Radiation Patterns at 5GHz and 6GHz when Fig. 4(b): Simulation Radiation Patterns at 5GHz and 7.2GHz when
Diode is in ON State Diode is in OFF State
IV. CONCLUSION [2] Nikolaou et al., "Pattern and frequency reconfigurable annular
slot antenna using PIN diodes," IEEE Trans. Antennas Propag.,
The idea of a simple microstrip patch antenna with vol. 54, no. 2, pp. 439, Dec. 2006.
multiple reconfigurable frequency band 5.01GHz, [3] G.H. Huff, J. Feng, S. Zhang, and J.T. Bernhard, "A novel
radiation pattern and frequency reconfigurable single tum square
7.2GHz and wide band from 4.3 GHz to 7.1 GHz is spiral microstrip antenna," IEEE Microwave and Wireless
proposed. By connecting and disconnecting the Components Letters, vol. 13, no.2, pp.57-59, 2003R. Nicole,
radiating element, shift of operating frequency of “Title of paper with only first word capitalized,” J. Name Stand.
antenna can be achieved with fixed radiation patterns. Abbrev., in press.
[4] J. Zhang, A. Wang, and P. Wang, "A Survey on reconfigurable
This method provides another approach for the antennas," ICMMT, vol. 3, 21-24, pp.1156-1159, 2008.
frequency-reconfigurable antenna design.
REFERENCES
[1] Upadhayay M.D. & Harish Kumar, Vikas K Rai, Leena
Varshney and Rajeev K Kanth “Multiband Planner Microstrip
Antenna” in Topical Conference on Antennas and Propagation
in wireless Communications (IEEE APWC 2011), Torino, Italy
Sept 12-16, 2011, PP.1360-1363. IEEE accession 978-1-4577-
0048-4/11J. Clerk Maxwell, A Treatise on Electricity and
Magnetism, 3rd ed., vol. 2. Oxford: Clarendon, 1892, pp.68-73.
Abstract—The design of multi band and switchable scheme in radio frequency identification (RFID)
circularly polarized Slotted Microstrip patch antenna is systems [7], and to increase the security complexity in
presented. The antenna is excited by Microstrip feed line. military wireless The dual band antenna is simply
The proposed antenna has square patch with the ring slot obtained by connecting the square and circular patch.
in the square patch, and the circular polarization
radiation can be generated by current perturbation due to
But still there is a need of higher information
diodes connected across the ring slot to the circular patch. transmission, for this purpose polarization switching
The p-i-n diodes across the ring slot alter the direction of concept is introduced. In polarization switching large
current path. Three p-i-n diodes simply switched the amount of information can be transmitted through a
circular polarization direction between the left handed single feed port without interference. This is achieved
and right handed. Linear polarization can also achieved if by orthogonal polarization modes that are left hand
both side ways diodes are OFF and centre diode is ON. circular polarization (LHCP) and right hand circular
Key parameters have been analysed on Electromagnetic polarization (RHCP). A switchable CP slot antenna
simulation software. The operating frequencies are with a p-i-n diode switch for different polarizations has
1.55GHz, 3.8GHz and 4.77 GHz and results obtained are
satisfactory.
also been reported [8]. It is also possible to achieve
switchable CP microstrip antennas with shorting walls
Keywords: Circular polarization (CP), microstrip sloted that are switched to the ground plane via p-i-n diodes
patch antenna, polarization switching, left hand circular [9], [10]. But by connecting the two patches at one
polarization (LHCP), right hand circular polarization more position will generate a new band without
(RHCP)
increasing the size of antenna.
In this paper a multi band polarization switchable
I. INTRODUCTION
antenna has been proposed. The structure is single feed
Aim of antenna design is to transmit large amount simple and compact producing multi band, circular and
of information to the longer distance, so the research in linear polarization. The orthogonal modes for circular
wireless communication is growing toward the polarization are generated by p-i-n diodes connecting
reduction of antenna size and get the higher rectangular and circular patch. The polarization
transmission bandwidth of antenna. Many applications switching is obtained by switching the p-i-n diodes.
require integrated multifunctional terminals [1], [2]. Different polarization has been discussed, which are
The concept of frequency reconfigurable and obtained by switching the diodes. In subsequent
polarization reconfigurable is developed. The pursuit of sections antenna structure and results have been
antenna banks with compact form factors has led to discussed. Section IV provides conclusion and possible
research into reconfigurable antennas with variable application of proposed antenna.
bandwidth [3], radiation characteristics [4], and
polarization [5]. And for achieving the multiple bands II. ANTENNA STRUCTURE
cascaded antennas are designed. In the cascaded
Fig.1 shows the geometry of the proposed
antenna large number of antennas is connected to form
multiband and dual switchable circular polarization
a single antenna and the switching is done to get the microstrip slotted patch antenna. Proposed antenna has
required frequency band. Such mobile terminals must three layer structures, bottom layer is a ground plane of
operate while acting as antenna banks because each thickness .01mm, on which there is a second layer of
service requires a different frequency band This dielectric substrate FR4 of thickness 2 mm and
structure increases the information rate but one dielectric constant is 4.9. Third layer is a patch which is
problem is also there that is size of antenna, to square shaped with a ring slot. The square patch length
implement large size of antenna in small devices is not is L mm. At the centre of the square patch there is a
possible, so the new research is done in reducing the ring slot of outer radius R2 and the inner radius R1.
size of antenna . Polarization diversity with switchable Fig.1 shows the construction of the square patch with a
CP is also used in order to reduce fading in wireless ring slot inside it. This ring slot separates the two
local area networks (WLANs) [6], as a modulation structures one is square patch and other circular patch
ISBN: 978-93-82880-20-2 11
Proceedings of National Conference on Trends in Signal Processing & Communication (TSPC’13) 12th–14th April 2013
which is inside the square patch. The separation III. IPLIMENTATION AND RESULTS
between the square patch and circular patch which is
inside it is 2mm or the width of ring slot is 2mm. This The proposed antenna is designed for multibands
proposed antenna produces linear polarization as well (1.55GHz, 3.8GHz and 4.77 GHz). The square patch
as circular polarization by introducing diode. (L=40mm) on the top of the FR4 substrate with Ɛr= 4.4
Fig.1 shows p-i-n diodes D1, D2 and D3. These p- and hight h=2mm. one ring slot in the square patch
i-n diodes providing the connection between the square with dimensions outer radius (R2=7 mm) and inner
patch and the circular patch. The diode D1 is connected radius (R1=5 mm), width of ring slot is 2mm, the three
between square patch and inner circular patch in left p-i-n diodes are located across the slot in order to
side, and diode D3 is connected between square patch control the current path. The λ/4 impedance
and inner circular patch in right side and diode D2 is at transformer is used for impedance matching between
down side. The polarization diversity is obtained by patch and 50Ω feed line.
switching different p-i-n diodes at different time. TABLE I shows the summary of the antenna
Linear polarization is obtained when P-I-N diode D2 is polarization. With the above said dimensions the
forward biased and D1 and D3 are reversed biased. To resonate frequencies for the linear polarization (when
get circular polarization two currents are fed at 900 diode D2 is forward biased and remaining two are
generating orthogonal components. The LHCP is reversed biased) are 2.1
obtained when left p-i-n diode i.e. D1 is forward biased TABLE I: SUMMARY OF ANTENNA POLARIZATION
along with D2, because of current direction. Similarly
RHCP is obtained when the right p-i-n diode (D3) is P-I-N diodes Biasing Polarization
D2 : ON
forward biased along with D2, reverse biasing the D1.
(Forward biased)
The antenna structure is three bands polarization D1,D3: OFF Linear Polarization (LP)
switching is obtained in only one band which is (Reverse biased)
generated by circular patch. It is seen that axial ratio D2,D3: ON
varied from 40 dB to 1 dB. (Forward Biased) Right Hand Circular
D1 : OFF Polarization (RHCP)
(Reverse Biased)
D1,D2: ON
(Forward Biased) Left Hand Circular
D3 : OFF Polarization (LHCP)
(Reverse biased)
GHz and 5.1 GHz. The band generated by the
circular patch can be switched to the circular
polarization by switching on either D1 or D3 we will
get three bands at this point. The bands generated by
circular patch can be RHCP or LHCP according to the
position of p-i-n diode forward biased.
Fig. 2 shows the simulated results of proposed
antenna when diode D1 and diode D3 are OFF and only
diode D2 is ON, in this case antenna radiate at two
frequency bands at 2.1GHz and at 5.1GHz. In this case
the proposed antenna is linearly polarized.
(a) (b)
(c)
Fig.
F 5: Simulated Farfield Pattern oof LHCP at (a) f=
=1.55GHz,
(b) f=
f 3.8GHz and (c)) f= 4.77GHz.
IV. CONCLU
USION
REFERENC
CES
[1] N. Jin, F. Yang,, and Y. Rahmat--Samii, “A novel patch antenna
with switchablee slot (PASS): D Dual frequency operation
o with
reversed circullar polarizationns,” IEEE Trans. Antennas
Propag., vol. 544, no. 3, pp. 1031––1034, Mar. 2006
6.
(c) [2] A. Calmon, G. G Pacheco, aand M. Teradaa, “A novel
reconfigurable UWB log-perioddic antenna,” in n Proc. IEEE
Fig. 4: Simulated Farfield Pattern of RHCPP at (a) f=1.55GH
Hz, Antennas Propag. Soc. Int. Symp., 2006, pp. 213– –216.
Hz and (c) f= 4.777GHz
(b) f= 3.8GH
Bhagwant Institute
e of Technologyy, Muzaffarnaga h, India 13
ar, Uttar Pradesh
Proceedings of National Conference on Trends in Signal Processing & Communication (TSPC’13) 12th–14th April 2013
[3] E. Erdil, K. Topalli, M. Unlu, O. A. Civi, and T. Akin, [7] M.-A. Kossel, R. Küng, H. Benedickter, and W. Bächtold, “An
“Frequency tunable microstrip patch antenna using RF MEMS active tagging system using circular-polarization modulation,”
technology,” IEEE Trans. Antennas Propag., vol. 55, no. 4, pp. IEEE Trans. Microw. Theory Tech., vol. 47, no. 12, pp. 2242–
1193–1196, Apr. 2007. 2248, Dec. 1999.
[4] X.-S. Yang, B.-Z.Wang, and W.Wu, “Pattern reconfigurable [8] M. K. Fries, M. Gräni, and R.Vahldieck, “A reconfigurable slot
patch antenna with two orthogonal quasi-Yagi arrays,” in Proc. antenn with switchable polarization,” IEEE Microw. Wireless
IEEE Antennas Propag. Soc. Int. Symp., 2005, vol. 2B, pp. Compon. Lett., vol. 13, no. 11, pp. 490–492, Nov. 2003.
617–620. [9] A. Khaleghi and M. Kamyab, “Reconfigurable single port
[5] Y.-F. Wu, C.-H. Wu, D.-Y. Sai, and F.-C. Chen, “A antenna with circular polarization diversity,” IEEE Trans.
reconfigurable quadri-polarization diversity aperture-coupled Antennas Propag., vol. 57, no. 2, pp. 555–559, Feb. 2009.
patch antenna,” IEEE Trans. Antennas Propag., vol. 55, no. 3, [10] W.-L. Liu, T.-R. Chen, S.-H. Chen, and J.-S. Row,
pp. 1009–1012, Mar. 2007. “Reconfigurable microstrip antenna with pattern and
[6] S.-T. Fang, “A novel polarization diversity antenna for WLAN polarisation diversities,” Electron. Lett., vol. 43, no. 2, pp. 77–
application,” in Proc. IEEE Antennas Propag. Soc. Int. Symp., 78, 2007.
2000, pp. 282–285.
ISBN: 978-93-82880-20-2 15
Proceedings of National Conference on Trends in Signal Processing & Communication (TSPC’13) 12th–14th April 2013
of microstrip antenna is narrow bandwidth, which is This E shaped patch antenna provides a 7.8 %
usually few percent (<1%). (2.34-2.53 GHz) impedance bandwidth with 5.8 dB
gain. The figure 3.2 (a) and 3.3 (a) is shown the return
B. Aperture Coupled Feed loss in dB and smith chat for equal size of slots in
In this the radiating patch is etched on the top of patch. The E shaped patch antenna is designed here,
the antenna substrate and feed line is etched on the has-33 dB return loss. The gain curve is shown in
bottom of substrate. The thickness and dielectric figure 3.4 (a).
constants of these two substrates may thus be chosen When the slot SL1 gets shorter, it forms the unequal
independently to optimize the distinct electrical slots E shaped patch antenna as shown in fig 3.1 [5].
functions of radiation and circuitry. The basic structure This unequal E shaped patch antenna provides circular
of aperture coupled microstrip antenna is shown in polarization. This antenna provides 16.4 % (2.15 GHz-
figure 2.2 [3]. 2.53
GHz) impedance bandwidth at-10 dB return loss
and 6dB gain. The figures 3.2 (b) and 3.3 (b) are shown
the return loss in dB and smith chat for unequal length
of slots in patch. The E shaped patch antenna designed
here has near about (-32) dB return loss. The gain curve
is shown in figure 3.4 (b). Table 2 shows simulated
results.
(b)
SL1
Fig. 3.2: Simulated Return Loss S11 V/S Frequency
W Ws TABLE 2: SIMULATED RESULTS FOR PROBE FEED PATCH ANTENNA
F
P Patch Frequency Gain Return Bandwidth
Shape Loss
E Shaped 2.45GHz 5.6dB -33dB 7.8%
Patch
SL2 Unequal 2.43GHz 6dB -31.2dB 16.4%
arms E
Shaped
Fig. 3.1: Geometry of Unequal E Shaped Patch Antenna Patch
(b) L
Fig. 3.3: Smith Chart for E Shaped Patch Antenna SL1
Lf
Ls
SL
2
W
Patch
Ground Plane
Ws
(b)
Fig. 4.2: Simulated Return Loss S11 V/S Frequency
Fig. 3.4: Simulated Gain Curve for E Shaped Patch
V. CONCLUSION
A small-size Microstrip antenna for WLAN
Application using probe feed and aperture feed is
proposed and successfully implemented. From the
comparative study of different configurations of
feeding techniques, it is concluded that unequal slot in
patch provides a bandwidth of 16.4 % with 6 dB gain
and an aperture coupled feed microstrip antenna
provides a bandwidth of around 24%.
This proposed microstrip antenna enhanced the
impedance bandwidth and provides good matching.
This antenna simulated by HFSS’11[7].
Fig. 4.3: Smith Chart for Aperture Coupled Antenna
REFERENCES
The resonance frequency of proposed antenna
occurs at 2.23 GHz. By changing the length of slots, [1] Antenna Theory, C.Balanis, Wiley, 2 nd edition (1997),
Chapter 14. ISBN 0-471-59268-4.
can change the resonant frequency. The best result for [2] D. M. Pozar, “Microstrip antenna aperture-coupled to a
the-10dB bandwidth is obtain with SL1 = SL2 = 16 mm microstripline,” Electron. Lett., vol. 21, no. 2, pp. 49-50, Jan.
and stub length Ls = 12.5 mm as shown in figure 4.2. 1985.
This aperture coupled antenna provides impedance [3] Michael Paul Civerolo, “Aperture Coupled Microstrip Antenna
Design and Analysis” Thesis, California Polytechnic State
bandwidth near about 24 %( at-10 dB return loss University; June 2010
graph). The minimum return loss is-40 dB at 2.23 GHz [4] Subodh Kumar Tripathi1, Vinay Kumar2, “E-Shaped Slotted
frequency. The figure 4.3 has shown smith chart for Microstrip Antenna with Enhanced Gain for Wireless
aperture coupled antenna. Table 4 shows all simulated communication” Internation-al Journal of Engineering Trends
and Technology-July to Aug Issue 2011
results. [5] Ahmed Khidre, Kai Fang Lee, Fan Yang, and Ate' Eisherbeni
TABLE 4: SIMULATED RESULTS FOR APERTURE COUPLED “Wide-band Circularly Polarized E-Shaped Patch Antenna for
FEED PATCH ANTENNA Wireless Applications” IEEE Antennas and Propagation
Magazine, Vol. 52, No.5, October 2010.
Patch Shape Frequency Return Loss Bandwidth [6] Qinjiang Rao, Tayeb A. Denidni, “A New Aperture Coupled
E Shaped 2.23GHz -40dB 24% Micro-strip Slot Antenna” IEEE Transactions on Antennas and
Patch Propagation, Vol. 53, No. 9, September 2005.
[7] Ansoft Designer, www.ansoft.com.
Abstract—A stacked notched rectangular microstrip mutually conflicting properties, that is, improvement of
patch antenna for WLAN and ISM band applications has one of the characteristics normally results in the
been presented in this paper. The desired band is degradation of the other. It is complex to incorporate all
obtained by tuning the dimensions of notches and size of the features such as uniformly high gain, broad
the stacked driven and parasitic patches. The same aspect
ratio is maintained for both the patches and their
bandwidth, efficiency, compact size, multi-resonant
respective notches. The bandwidth and gain of this characteristics along with the desired radiation pattern
antenna is 23% and 7.87dBi, respectively. Antenna into a single antenna design. In the present work, a high
efficiency of 87% is achieved in the designed antenna. The gain, broad bandwidth, high efficiency (87%)
bandwidth, gain, size reduction and efficiency obtained in diagonally probe fed stacked notched microstrip patch
the designed antenna makes it a potential radiator for antenna has been proposed and discussed. Rectangular
various applications. notch has been incorporated along the non-radiating
Keywords: Microstrip patch antenna (MSA), Wireless edge of both the lower and upper parasitic patch, an air
local area network (WLAN), ISM band, Wideband antenna, gap has been provided in between two layers. Air gap
Stacked notched rectangular microstrip patch antenna enhances gain11. This antenna has been designed for
WLAN applications at the IEEE 802.11a bands (5.15–
I. INTRODUCTION 5.35 GHz and 5.725 –.825 GHz) and also for 5-6 GHz
ISM bands. The bandwidth of proposed is 4.85GHz and
The demand for compact and multifunctional 6.07 GHz and also the resonating frequency of this
wireless communication systems has spurred the antenna is 6.07 GHz.
development of miniaturized broadband antennas with
high gain. Recent advancement in high speed WLAN
II. ANTENNA DESIGN
and similar applications need the antennas of broad
bandwidth, small size, high gain, and enhanced Figure 1 depicts the cross sectional view of the
efficiency and these antennas are also used in ISM stacked-notched rectangular microstrip patch antenna.
bands. Microstrip patch antennas are widely used in The proposed antenna comprises two patches: lower
these applications due to advantageous features like rectangular patch and upper rectangular patch element
compactness, light weight and high efficiency. as a parasitic element which are stacked by separating
However, one of the major limitations of the microstrip with an air gap. Figures 2(a) and 2(b), respectively
patch antenna is the narrow bandwidth1-4. There are show the geometrical structures of the driven and
well-known established techniques to enhance the parasitic rectangular microstrip patches of the proposed
bandwidth of these antennas2-10 such as increasing the antenna.
substrate thickness using low dielectric substrate; The geometrical dimensions of the lower that is
incorporating various impedance-matching and feeding driven rectangular microstrip patch are length Ll =
elements using multi resonators, multi-slot hole- 18.14 mm and width Wl = 14.4 mm. The rectangular
coupling6 using planar parasitic elements7, notch of dimensions: length a = 3.17 mm and width b =
incorporating two thin substrates8, creating multi- 2.5 mm, respectively has been created on the non-
couple staggered slots9, stacking of substrates10, etc. radiating edge of the driven patch. The lower patch is
Shigeru Egashira and Eisuke Nishiyama reported a designed on RT Duriod substrate of thickness h1 =
stacked microstrip patch antenna with two parasitic 1.575 mm, dielectric constant εr = 2.2 and loss tangent
elements. The first parasitic enhances the bandwidth = 0.0009. The length and width of upper rectangular
and gain has been increased due to second parasitic parasitic patch is Lu = 19.15 mm and Wu =15.2 mm,
element. Nasimuddin et al. presented the circularly respectively. The length and width of the notch on the
polarized stacked microstrip antenna to improve the upper parasitic patch are: c = 2.52 mm and d = 2 mm,
axial ratio bandwidth. Notch loading in a microstrip respectively. In widths of both the patches, the
patch antennas is used to reduce the impedance respective notch has been cut at an identical distance
mismatch and to enhance the bandwidth. However, the from the length edges. The RT Duriod substrate of
bandwidth, gain and size of an antenna are generally thickness h2 = 0.787 mm, dielectric constant εr = 2.2,
ISBN: 978-93-82880-20-2 19
Proceedings of National Conference on Trends in Signal Processing & Communication (TSPC’13) 12th–14th April 2013
and loss tangent = 0.0009 has been used for the upper developed in the RLC circuit of notched parasitic
patch. Both the patches are separated by an air gap of g element. These two RLC circuits get capacitively
= 1.4 mm. In the proposed design, the lower coupled due to the air gap. This lowers the upper
rectangular patch is co-axially excited at the location x resonant frequency of the dual band at good matched
= 4.65 mm and y =-5.86 mm with the feed radius of 0.4 condition and the broadband performance is achieved.
mm. The method of moments based IE3D Thus, by optimizing the dimensions of both the
electromagnetic simulator has been used to design and rectangular patches with the respective notches and by
simulate this antenna. Figure 3 shows the simulated adjusting the air gap between them, broadband
structure of stacked-notched rectangular microstrip performance can be achieved over the frequency band
patch antenna. of interest.
Abstract—The design of a ultra wide band microstrip centre strip d are 0.8mm and 2.4mm respectively. The
antenna is reported here. The antenna is providing three length of the patch Lp is 37.5mm. The fringing length
UWB compatible bands in the frequency range of 7.24- ΔL=0.7mm. This antenna designed on Rogers3010 (tm)
7.55GHz and 8.27-8.87GHz and 9.27-9.93GHz with the substrate with dielectric constant 10.2, loss tangent
bandwidth of 310MHz, 500MHz & 660MHz respectively
and their corresponding return loss of 29.53dB, 32.6dB
0.0035 and height 1.5mm. The feeding is applied
and 29.7dB. The simulation is performed using HFSS through a 50Ω strip by inset feeding technique.
software. Antenna is designed using Rogers3010 (tm)
substrate and with the substrate dimensions of
37.5x24mm. Antenna can be used for UWB
communication in the frequency range of 7GHz–10GHz.
Keywords: Microstrip patch, inset feed, multiband
antenna, UWB.
I. INTRODUCTION
In recent years, microstrip patch antennas are
becoming attractive for use in wireless communication
systems [1] [7] including ultra wideband (UWB)
systems due to their attractive merits of simple
structure and low profile. Ultra-wideband
communications is fundamentally different from all
other communication techniques because it employs
extremely narrow RF pulses to communicate between
transmitters and receivers. Planar monopole antennas
are able to radiate bi-directional radiation patterns with
larger bandwidths. Patch can take various
configurations such as rectangle, circular, diamond etc.
Inset feeding is supposed to be better candidates
because of their simple configuration, manufacturing Fig. 1: Geometry of Antenna
advantages, repeatability and low cost. Inset-fed
antennas have been studied extensively [2-6]. In III. RESULTS AND DISCUSSIONS
microwave and wireless communication, there is
requirement of components having light weight, high The designed antenna is radiating in three bands
performance, easier fabrication, more reliable and which are lying in ultra wide band frequency range of
smaller size, although reduction in size of an antenna 3.1-10.6GHz. The radiating bands are 7.24-7.55GHz,
depends on antenna constraints which cannot be 8.27-8.87GHz and 9.27-9.93GHz and their
changed easily so we have to make a tradeoff between corresponding return loss are-29.63dB,-32.6dB,-
them. The antenna is designed considering these 29.7dB. The antenna is designed to resonate at 7.5
tradeoffs. GHz. The return loss plot of antenna at this resonant
frequency is shown in figure 2. Figure 3 is showing the
II. DESIGN PARAMETERS AND STRUCTURE ANALYSIS VSWR plot of the antenna. The theta and phi radiation
plot of the antenna with respect angle is shown in
The antenna geometry is shown in figure.1. The figure 5 & 6 respectively. The 3D rectangular and polar
total length of the antenna L is 38.9mm and the width field pattern with respect to phi and theta at 7.5GHz is
of antenna W is 24mm. The gap c and the width of the shown in figure 8 & 9 respectively. For the resonance
ISBN: 978-93-82880-20-2 23
Proceedings of National Conference on Trends in Signal Processing & Communication (TSPC’13) 12th–14th April 2013
Solution
Frequency Bandwidth Return
Frequency
Range(GHZ) (MHz) Loss (dB)
(GHz)
7.24-7.55 310 MHz -29.53 dB
7.5 GHz 8.27-8.87 500 MHz -32.6 dB
9.27-9.93 660 MHz -29.7 dB
REFERENCES
[1] J.R. James and P.S. Hall, Handbook of Microstrip Antennas,
Fig. 8: 3D Rectangular Radiation Pattern of Antenna vol. 1, Peter Peregrinus Ltd., London, 1989.
[2] R Garg, P. Bhartia, I. J. Bahl and A. Ittipiboon, Microstrip
Antenna Design Handbook, Artech House, 2001.
IV. CONCLUSION [3] D. M. Pozar, “Microstrip antennas”, IEEE Transaction on
Antenna and Propagation, Vol. 80, pp. 79–91, Jan. 1992..
A microstrip antenna is designed for ultra wide [4] R. Waterhouse, “Small microstrip patch antenna,” Electron.
band applications, which is providing excellent Lett., Vol.31, pp. 604–605, 1995.
radiation efficiency of greater than 95% and the [5] Constantine A. Balanis, Antenna theory-Analysis and design,
bandwidth efficiency of 49%. The antenna is providing 2nd edition, John Wiley & sons, Inc., pp. 727–735, 1997.
[6] Anders G. Derneryd, “A Theoretical Investigation of the
three bands having the bandwidth of 310 MHz, Rectangular Microstrip Antenna Element.” IEEE Transcation
500MHz and 660 MHz with their corresponding return on Antennas and Propagation, Vol. AP-26, No. 4, pp. 532–535,
loss are-29.63dB,-32.6dB,-29.7dB. This is satisfying July 1978.
the bandwidth requirement of Ultra wideband [7] Thomas A. Milligan, Modern Antenna Design, 2nd Edition,
John Wiley & Sons, Inc., chapter 5&6, 2005.
communication. [8] Ansoft HFSS 14, Ansoft Corporation.
Abstract—This paper describes the design and approximately half wavelength at the center frequency
development of band pass filter (BPF) at 94 GHz. This [1]. Now Fig.1 (b) is redrawn with negative length
design uses capacitively coupled series resonators in transmission line sections on either side of series
suspended stripline configuration. This design is modeled capacitors, which forms the equivalent circuit of an
and optimized using CST (Computer Simulation
Technology) microwave studio. The 94 GHz band pass
admittance inverter as shown in Fig.1 (c). The length
filter exhibits an insertion loss of 1 dB with 3 dB relative ‘φ’ is λ/2 at fo. Electrical length θi of the ith section is
bandwidth at a center frequency of 94 GHz and the given by;
ɸ ɸ
return loss is better than-18 dB at a center frequency. The θº= π+ + for i=1, 2…….N (1)
designed and fabricated 94 GHz band pass filter shows The capacitive gap coupled filter can be modeled
the good performance for planar integrated millimeter-
wave circuits.
as shown in Fig.1 (d). Procedure to get electrical length
of resonating section length has been carried out in an
Keywords: BPF, End coupled, Suspended stripline, innovative way which is very accurate at 94 GHz
Transition frequency. We need to derive [ABCD] and [S] matrix
for J and K inverters which is used to calculate filter
I. INTRODUCTION parameters.
Low cost compact filters are important aspect of The end coupled suspended stripline band pass
millimeter wave system. Almost all receiver systems filters are most promising due to its compact size,
require input filter for channelization of input signal. lightweight, low cost and ease of fabrication. In E plane
The basic objectives in any filter design is that of circuit supporting dielectric (e.g. Fin line) causes
achieving low insertion loss and VSWR within the additional losses. It may therefore often be
required bandwidth while attaining the required out of advantageous to restrict the design for high Q
band signal rejection [1]. Due to very less channel millimeter wave circuit to pure metal inserts placed in
space available at 94 GHz, filter has been designed the E-plane of rectangular waveguide without any
using capacitively coupled series resonators in substrate. The design specification for the above filter
suspended stripline configuration as shown in fig.1. is given below.
1. Center frequency: 94 GHz
2. Bandwidth: 6 GHz
3. Attenuation at ±5 GHz from center: 30 dB
4. Insertion loss: < 2 dB
5. Return loss: > 15dB
6. Ripple: 0.1 dB
7. Prototype: Tchebyscheff
26 ISBN: 978-93-82880-20-2
Design and Development of Band Pass Filter at 94 GHz
V. CONCLUSION
The present filter was tested from 83 GHz to 105
GHz. Suspended stripline filter has 1.5-2.5 dB insertion
loss over a bandwidth of 6 GHz excluding transition
loss. The deviation observed was mainly attributed to
the mechanical housing, whose precision is a very
important aspect at 94 GHz.
REFERENCES
[1] David M. Pozar, “Microwave Engineering”, Third Edition,
Fig. 4: CST Model of 5 Section Band Pass Filter John Wiley & Sons, Inc., USA, 2005.
[2] G.Matthaei, L. Young & E.M.T.Jones, “Microwave Filters,
Impedance Matching Networks, and Coupling Structures”,
Artech House.
[3] Richard M. Dougherly “MM-Wave filter design with suspended
stripline.”Microwave Journal July1986. pp 75-84
[4] L.Q.BUI, D BALL “Broad-Band Millimeter –wave E-plane
Band pass filters” IEEE MTT-32 Dec 1984 pp 1655-1658.
[5] Y. Konishi and K. Venekada “The design of a band pass filter
with inductive strip planner circuit mounted in waveguide”
IEEE Trans. Microwave Theory Tech, Vol MTT-22 pp 1209-
1216 1974.
Abstract—A novel compact reconfigurable multiband diode, the connectivity is made open between two
antenna is proposed. The antenna is designed with the diode points. The designed antenna operates in four
40mm x 40mm substrate with the dielectric constant of bands which are tunable in the range of 8 to19GHz.
10.2 and the thickness of 1.27mm. The substrate used is
Roger RT/Duroid 6010. By achieving the switching
multiband functionality, this antenna is almost covering
the complete range of Ku-band with the radiation
efficiency >85 % for all the switching combinations. The
operating frequencies bands are Band 1 (13.7-14.1GHz),
Band 2 (14.15-14.55GHz), Band 3 (14.4-14.6GHz), Band 4
(18.1-18.9 GHz).
Keywords: Reconfiguration antenna; Slot antenna;
Probe feed antenna
I. INTRODUCTION
Reconfigurable antenna reveals various beneficial
aspects as compared to the conventional antenna at the
same class. It achieved multi function for a similar
antenna. In current scenario of antenna development,
the need of small size multifunctional antenna is on the
rise. However the size reduction causes the poor
antenna performance such as low gain and efficiency
[8], [9]. The advantage of reconfigurable antenna is that
the antenna volume can be reused by reconfiguring the
antenna to operate in different or multiband; thus
reducing the overall size. A reconfigurable antenna can
be used for various surveillance and military
communication by dynamically selection of bands via
RF switches such as PIN or Varactor diode. For
simulation, PIN diode can be modeled using RLC sheet
[1]. When diode is in forward or ON state, it behaves Fig. 1: Proposed Structure of Antenna Dimensions: L=W=40mm,
like a resistor and in OFF state, work as a capacitor. In Wp = Lp = 38 mm, s=11.7mm, g=14mm, r=3mm,
a next generation wireless system, wide band antenna is width of the slot is 1.5mm
required to transmit multiple data, an example of such
application is an MB-OFDM ultra wide band system II. ANTENNA ANALYSIS AND DISCUSSION
consists of 14 channels with a bandwidth of 528 MHz This paper presents a novel small size antenna with
[12]. A reconfigurable antenna has more advantages in reconfigurable multi-bands. The presented antenna is
terms of increasing antenna capabilities by tuning its designed on Roger RT/Duroid 6010 substrate with the
performances [2-7]. The dynamic properties of dielectric of 2.2 and the thickness of 1.27mm and the
reconfigurable antenna can be achieved by changing tangent loss of 0.0023. Antenna consists of 8 switches.
the state of to either ON or OFF of the RF switches For all the switch combinations, the return loss S11 is <
such as PIN diodes. This will eliminate the need to -10dB as specified in Table1. Antenna is designed for
increase the dimensions or coverage area. To achieve the center frequency of 14.1GHz and antenna is having
switching, in forward mode, the strip of equivalent good characteristics for 13.5 to 15 GHz range.
resistance of diode is used. For negative cycle of the
ISBN: 978-93-82880-20-2 29
Proceedings of National Conference on Trends in Signal Processing & Communication (TSPC’13) 12th–14th April 2013
Dimension wise, the proposed 40 x 40mm antenna Switch Position Frequency Bandwidth Return
is small as compared to the traditional ones. Fig. 1 Range (GHz) (MHz) Loss (dB)
All switches are
shows the schematic of antenna. This antenna is excited ON (fig.2 a)
18.1-18.9 800 MHz -20.1 dB
by probe feed at the center of the modified patch. the All switches OFF 14.4-14.6 200 MHz -28.3 dB
switching configuration of the antenna is shown in (fig.2 b) 13.7-14.1 400 MHz -19.2 dB
figure 2. The maximum return loss of the antenna is - Switches 1,2,5 &
6 are OFF (fig.2 14.15-14.55 400 MHz -32 dB
32dB in the frequency band of 14.15-14.55GHz as c)
shown in figure 5. The widest bandwidth of the antenna
is 800MHz with a return loss of -20.1dB in the
frequency range of 18.1-18.9GHz when all the switches
are ON (ref fig 2(a)). When all the switches are OFF,
antenna is operating in two bands 14.4-14.6GHz and
13.7-14.1GHz with the corresponding return loss of -
28.3dB and -19.2dB as shown in figure 5. The radiation
patterns of the antenna for the different switching
combinations are shown in fig. 6-8. The 3D rectangular
radiation pattern of antenna is shown in figure 9. For all
the switching combinations, the efficiency ranges from
0.85 to 0.90. figure 10 shows the simulated antenna Fig. 3: Return Loss Plot of Antenna (for Figure 2a)
radiation efficiency graph.
III. CONCLUSION
A novel small size reconfigurable multiband
antenna is designed and simulated using HFSS
software [13]. The presented antenna is having the
radiating range of Ku-Band (12-18GHz). The operating
frequency ranges are providing the acceptable
reflection coefficient, s11< -10dB, the VSWR ranging
from 1.21-1.93(VSWR< 2) in the operating bands and
the efficiency > 85 % for all the switching
combinations. However, the need of this structure is to
optimize and removal of the notch-bands to expand the
radiating band. Our future work is to optimize the
antenna performance with fewer switches and to
remove the undesired notch-bands of the antenna.
[7] M. Jusoh, M. R. Kamarudin, M. F. Jamlos, M. F. Malek, “A [10] Constantine A. Balanis, Antenna theory- Analysis and design,
Novel Compact Reconfigurable Multi-Band Antenna,” IEEE 2nd edition, John Wiley & sons, Inc., pp. 727–735, 1997.
international RF and Microwave Conference (RFM 2011), [11] Thomas A. Milligan, Modern Antenna Design, Second Edition,
Seremban, Malaysia 2011. John Wiley & Sons, Inc., chapter 5&6, 2005.
[8] A. Skrivervik, J.-F. Zurcher, S. O, and J. Mosig, “Pcs antenna [12] Lee, W., Kim, H., and Yoon, Y.J.: “Reconfigurable slot antenna
design: the challenge of miniaturization,” IEEE Antennas with wide bandwidth‟. IEEE Antenna and Propagation Society
Propagat. Mag., vol. 43, no. 4, pp. 12–27, 2001. Int. Symp., July 2006, Vol. 4, pp. 3063–3431.
[9] R. Azadegan and K. Sarabandi. “A novel approach for [13] Ansoft HFSS 14, Ansoft Corporation.
miniaturization of slot antennas,” IEEE Antennas
Propagat.Mag., vol. 51, no. 3, pp. 421–429, 2003.
A Reconfigurable Multiband
Square Patch Antenna
Shweta Gautam1, AjitYadav2 and Mithilesh Kumar3
1,2,3
Electronics Department, University College of Engineering,
Rajasthan Technical University, Kota
e-mail: 1s.ecom.og@gmail.com, 2ajity2@gmail.com, 3mith_kr@yahoo.com
Abstract—The design of simple reconfigurable diodes, which provide different configuration and
multiband square patchantenna is proposed, the antenna different shorting ways between the patch and the
is excited by Microstrip feed line. The proposed antenna
ground. This antenna works satisfactory for single
has square patch that is shorted to the ground plane
through six shorting walls with six diodes. Three stubs band, dual band and triple bands.
with three diodes in each side of the square patch are
present. The p-i-n diodes provide the opening and II. ANTENNA STRUCTURE
shorting of ground with the patch by taking the different
combination of shorting diodes. Designed antenna radiate Fig.1 shows the geometry of the proposed
at a number of frequency bands, it shows single band, reconfigurable multiband square patch antenna.
dual band and multi band operation. Key parameters Proposed antenna has three layer structure, bottom
have been analysed on Electromagnetic simulation layer is a ground plane of thickness .01mm, on which
software. The proposed antenna shows different dual there is a second layer of dielectric substrate FR4 of
band operations and multi band operation. Dual band
thickness 3 mm and dielectric constant is 4.9. The
occurs satisfactorily at 3.57GHz and 6.2GHz. Another
dual band occurs satisfactory at 1.76GHz and 6.2GHz, length of the ground plane of 70mm and the width is
and a multi bands occur at 3.06, 4.4 and 6.65 GHz. also 70mm. FR4 substrate too has 70 mm length and 70
mm width. The third layer is a square patch of length L.
Keywords: shorting wall, shorting p-i-n,
The square patch with the shorting walls is shown
multibandsquare patch, frequency reconfigurable antenna,
dual band. in Fig1 (b) and fig1(c).There are six shorting walls 3 on
each vertical sides of the square patch. These shorting
I. INTRODUCTION walls are not directly shorting the ground plane to the
patch, diodes are connected to provide desirable
In wireless communication patch antenna radiates shorting between ground plane and the square patch.
for narrow bandwidth which limits its application [1]. Fig.1 (b) shows the left view of the proposed
So today’s requirement is small size simple multiband antenna.There are three shorting walls in left side of the
or wide band antennas which are convenient to patch Shorting wall 1, shorting wall2 and shorting wall
integrate in communication system. Wide band 3.and three p-i-n diodes D1, D2 and D3 are also shown
operation of antenna can be achieved by doing in Fig.1 (b). Bottom of the shorting wall 1 is connected
modification in structure of antenna.Some techniques to the ground plane but top is not connected directly to
are shorting wall [2], folded shorting wall [3], u-slot the patch, top of shorting wall 1 is connected to the
array [4], slots form [5],y-v slot [6], stacked patch [7], positive of the p-i-n diode D1 and negative of the p-i-n
pair of slits on the patch (with total size of the antenna diode is connected to the patch. In such a way top of
2
150* 150 mm ) [8], E-shaped patch on thick substrates the shorting wall 2 is connected to the patch and its
with ground plane size of 140 *210 mm [9] and using bottom is not connected directly to the ground plane,
circular arc shaped slot on thick substrate[10]. there is a p-i-n diode D2 between the bottom of the
This paper presents the frequency reconfigurable shorting wall 2 and the ground plane. Similarly
dual band and multi band operation of proposed shorting wall 3 is also shown in Fig.1 (b). Top of the
antenna. This proposed antenna is multi tasking shorting wall 3 is connected to the patch via p-i-n diode
antenna with different frequency of operation. For D3.
achieving multi band operation patch structure is The same structure is replicated in the right view
modified. The proposed antenna has square patch with of the proposed antenna shown in Fig.1(c) shorting
microstrip feed line. This patch structure is modified by wall 4 andshorting wall 6 are connected to the patch via
introducing the shorting wall in vertical sides of the diodes D4 and D6 respectively. And shorting wall 5 is
patch which provide shorting between the patch and the connected to the ground plane via p-i-n diode D5.
ground. This shorting is achieved with help of p-i-n
ISBN: 978-93-82880-20-2 33
Proceedingss of National Co
onference on Tre
ends in Signal Processing
P & Co
ommunication (TSPC’13) 12th–
–14th April 2013
3
34 Bhagw
want Institute of Technology, Muzaffarnagar, Uttar
U Pradesh, In
ndia
A Reconfigurable Multiband Square Pa
atch Antenna
IV
V. IPLIMENTA
ATION AND RESULTS casee antenna woork as single band antennaa shown in
Fig.4 and radiate at frequency 33.13GHz.
Propossed antenna has square patch of leength
Case-5, wheen all diodes are ON agaain antenna
L=20mm, FR4 substratte of εr=4.9, and thicknesss of
radiate triple bannd shown in 33.06GHz, 4.4 GHz, 6.65
3mm, lenggth of 50Ω microstrip
m line is 10mm andd λ/4
GHzz. Some otherr cases are alsoo possible.
quarter waave transformeer length is 155mm to matchh the
feed line too the patch. Shhorting walls length
l is 2mmm and
width is 1.5mm. The patch and the ground have
thickness of
o .01mm.
The proposed antennna work as single band, dual
band and multi
m band anntenna by shoorting the diffferent
shorting walls
w via p-i-n diodes .A feww cases are stuudied
and shownn in this paper..
Case-11, when dioddes D1, D5 are a ON and other o
diodes are OFF in this case
c antenna work
w as dual band
antenna annd radiate at frequencies
f 3
3.57 and 6.2 GHz.
G
Fig.2 showws the s-param meter verses frequency
f curvve of
Case-1.Casse-2, when dioded D2 andd D5 are ON N and
other diodes are OFF, proposed
p anteenna still worrk as
dual band antenna but two frequenncies are diffferent
from case-1 that is 1.76 and 3.32 GHzz.
Bhagwant Institute
e of Technologyy, Muzaffarnaga h, India 35
ar, Uttar Pradesh
Proceedings of National Conference on Trends in Signal Processing & Communication (TSPC’13) 12th–14th April 2013
REFERENCES
[1] G. Kumar and K. P. Ray, Broadband MicrostripAntennas.
Boston, MA: Artech House, 2003, pp. 18–23.
[2] C. Mak, R. Chair, K. Lee, K. Luk, and A. Kishk, “Half U-slot
patch antenna with shorting wall,” Elect. Lett., vol. 39, pp.
Fig. 6: Simulated Far Field Pattern of Case-2, at 1.76GHz 1779–1780, 2003.
and 3.32GHz [3] Y. Li, R. Chair, K.M. Luk, and K. F. Lee, “Broadband
triangular patch antenna with a folded shorting wall,”
The Fig.5 shows the far field pattern of case-1 [4] IEEE Antennas Wireless Propag. Lett., vol. 3, no. 1, pp.189–
shown in Table-II. Far field pattern are studied at 192, Dec. 2004.
[5] H. Wang, X. B. Huang, and D. G. Fang, “A single layer
3.57GHz and 6.2GHz, at frequency 3.57GHz the main wideband U-slot microstrip patch antenna array,” IEEE
lobe magnitude is 6.8dBi and main lobe direction is 5.0 Antennas Wireless Propag. Lett., vol. 7, pp. 9–12, 2008.
degree .At frequency6.2GHz the main lobe magnitude [6] Y. Lee and J. Sun, “A new printed antenna for multiband
is 5.8 dBi and main lobe direction is 40.0 degree. Fig .6 wireless applications,” IEEE AntennasWirelessPropag.Lett.,
vol. 8, pp. 402–405, 2009.
shows the far field pattern of case-2 shown in Table-II. [7] S. Qu and Q. Xue, “A Y-shaped stub proximity coupled V-slot
Due to dual band operation far field pattern are studied Microstrip patch antenna,” IEEE AntennasWireless Propag.
at 1.76GHz and 3.32 GHz. At 1.76GHz the main lobe Lett., vol. 6, pp. 40–42, 2007.
magnitude is 4.4dBi and the main lobe direction is at [8] J. Anguera, C. Puente, C. Borja, and J. Soler, “Dual-frequency
broadband-stacked microstrip antenna using a reactive loading
20.0 degree .and at 3.32GHz the main lobe magnitude and a fractalshaped radiating edge,” IEEE Antennas Wireless
is 6.8dBi and the main lobe direction is 0.0 degree. Propag. Lett., vol. 6, pp. 309–312, 2007.
[9] K.-L. Wong and W.-H. Hsu, “A broad-band rectangular patch
V. CONCLUSION antenna with a pair of wide slits,” IEEE Trans.Antennas
Propag., vol. 49, no. 9, pp. 1345–1347, Sep.2001.
In this letter, a simple reconfigurable multiband [10] F.Yang, X. Zhang, X.Ye, and Y. Rahmat-Samii, “Wide-band E-
shaped patch antennas for wireless communications,” IEEE
square patch antenna is proposed. The dual band and Trans. Antennas Propag., vol. 49, no. 7, pp. 1094–1100, Jul.
multi band operation of antenna is taken by shorting the 2001.
ground plane and the patch through shorting walls and [11] R. Bhalla and L. Shafai, “Broadband patch antenna with a
p-i-n diodes. By making the diodes ON and OFF circular arc shaped slot,” in Proc. IEEE AntennasPropag. Soc.
Int. Symp., 2002, vol. 1, pp. 394–397.
different configurations are taken and due to these
Abstract—In this paper the UWB technology- UWB system must use a band-notch filter to prevent
(Operating in broad frequency range of 3.1–10.6 GHz) interference with WLAN systems.
based dual notch WLAN notch has shown great Of course, adding a filter increases UWB system
achievement for high-speed wireless communications. To complexity ideally, the notch can be included in the
satisfy the UWB system requirements, a band pass filter
with a broad pass band width, low insertion loss, and high
antenna. Various methods can achieve an antenna with
stop-band suppression are needed .UWB filter with band-notched characteristics. Conventional approaches
wireless local area network (WLAN) notch at 5.6 GHz a including cutting a slot (such as a U-C-or V-shaped
compact UWB antenna with coplanar-waveguide (CPW) slot) into the antenna patch[9-11]. While this
feed line was developed with integral notch at 5.0 to 6.0 technique typically achieves a notch for one band, it
GHz to eliminate interference from WLAN signals. The doesn’t prevent interference on other narrow bands.
compact antenna consists of a 28 x 30 mm patch with a C- more recently, antennas with two notched bands have
shaped slot in the radiating element to achieve the notch been proposed[12-15]. But most of these notched
which was measured 28 x 30 mm printed on FR-4 antennas are large in size and not suitable for compact
substrate. The circuit-board material has 1.6-mm
thickness, relative permittivity of 4.4, and loss tangent tan
wireless designs. As an alternative to these larger
δ of 0.02 antenna designs, a compact UWB antenna was
developed that measures just 28 x 30 mm. By etching a
I. INTRODUCTION C-shaped slot in the radiating element, a single band-
notched characteristic from 5 to 6 GHz was obtained.
Federal Communications Commission’s (FCC’s) And by etching two nested C-shaped slots in the
approved ultra wideband (UWB) frequency range from radiating patch, dual band-notched characteristics can
3.1 to 10.6 GHz Antennas with multiple band stop be achieved for the compact UWB antenna [14].
characteristics can be useful for multiple-network Rectangular patch antennas are typically known to
applications. by inserting C-shaped slots in various provide narrowband characteristics. To improve the
antenna configurations, it has proven possible to create operating bandwidth, the authors shaped the bottom of
a desired frequency stop band[1] Such antennas have the antenna patch into an arc. In practical applications,
value for selected bands within the Antennas are the size of ground plane is finite and the direction of
critical components in UWB systems that must be maximum radiation tilts somewhat upward from the
carefully designed for stable radiation patterns and horizontal plane. To reduce this beam tilting, the
polarization as well as good matching and low ground plane of the proposed antenna was designed for
dispersion. The current researchers discovered that by a rounded rather than a rectangular shape. The arc-
etching a C-shaped slot in the radiating element, a shaped patch and tapered ground plane make good
compact UWB antenna can be fabricated with strong broadband impedance matching of the antenna
band-notched characteristic from 5.0 to 6.0 GHz. possible. To further expand the impedance bandwidth,
In recent antenna developments, planar printed several forms were etched from the patch: a right-angle
antennas fed with microstrip or coplanar-waveguide triangle on the upper corners of the patch and a small
(CPW) transmission lines have received a great deal of fan angle on each side of the ground plane near the
attention for their high radiation efficiencies in compact feed line. These improvements are highlighted in Fig.
sizes, as well as their ease of integration with other 1, Figure 2 shown a comparison of the simulated
circuits. CPW-fed antennas are appealing because their reflection coefficient (S11) for an antenna with full
feed lines and slots are on one side of the substrate[2] rectangular radiating patch with an antenna modified on
For wideband coverage, different patch geometries the bottom with cuts on the upper corners and the small
have been developed including X-shaped[3] fork- angle on each side of the ground plane when all the
shaped, square-shaped, elliptical-shaped spade-shaped, dimensions are the same. The first line demonstrates
and circular-shaped patches[4-5] Because the 5.150-to- the rectangular patch antenna has a narrowband
5.825-GHz band is limited by IEEE 802.11a for characteristic. The second line shows the impedance
wireless-local-area-network (WLAN) applications, any bandwidth is broadened when the bottom of the patch
ISBN: 978-93-82880-20-2 37
Proceedings of National Conference on Trends in Signal Processing & Communication (TSPC’13) 12th–14th April 2013
is shaped into an arc. On the basis of this, the third 10.7 mm, L7 = 5.4 mm, L8 = 12.9 mm, R = 15 mm, g
line about the upper corners modification indicates that =0.7 mm, f = 3.6 mm, c = 0.3 mm, t = 0.3 mm, and H =
the bandwidth has been further expanded compared 1.5mm. By selecting optimal parameters, the proposed
with the second line the main advantages that this antenna can be tuned to operate across the UWB band.
modification is reducing the lowest frequency The Since UWB radios share part of the spectrum with
fourth line demonstrates the small angle on each side of HIPERLAN/2 applications (5.15 to 5.35 GHz and
the ground plane near the feed line will affect the 5.470 to 5.725 GHz) and WLAN applications using
characteristic impedance of the CPW line and also the IEEE 802.1la protocol (5.15 to 5.35
contribute to the modified antenna’s very wide GHz and 5.725 to 5.825 GHz), an UWB antenna
impedance bandwidth[12]. with reconfigurable band-rejection characteristic at
WLAN frequencies is highly desirable. Several designs
of UWB antennas with band-rejection characteristics
have been investigated and successfully implemented
in the past. The simple and commonly used approach
is to incorporate slots into the antennas main radiator.
Figure 4 shows a schematic diagram of the UWB
planar antenna (referred to as antenna B) with
filtering property operating in the 5-to-6-GHz band.
Figure 5 shows the current distribution around the slot
of antenna B at the 5.5-GHz notch frequency. At 5.5
GHz, the current flows mainly around the C-shaped
slot. The impedance is nearly zero at the top of the slot
Fig. 1: Show the Process of Improving a Patch Antenna and the impedance is very high near the antenna
with Built-in Notches feed line. In this case, the high impedance at the feed
point leads to the desired impedance mismatch near the
5.5-GHz notch frequency. In addition to WLAN
systems, WiMAX (3.3 to 3.6 GHz) and C-band
communications systems (3.7 to 4.2 GHz) may
interfere with UWB communications systems,
prompting the need for antennas with multiple notched
bands. Antennas with dual notched bands can be
realized by using two nested C-shaped slots in the
radiating patch (antenna A), which can yield band-
notched characteristics centered at 3.4 and 5.5 GHz,
respectively[14]. Figure 6 shows the geometry of the
UWB antenna (referred to as antenna C) with dual
notched-band characteristics. Since mutual coupling
Fig. 2: Curves Compare the Simulated VSWRs of Antennas A and B
exists between the exterior and interior C-shaped slots,
the HFSS simulated values did not match the
II. MESUREMENT & SIMULATION
predicted values. The optimized design parameters
Figure 3 shown the final geometry of the proposed are R1 =4.5 mm, R2 = 4.2 mm, R3 = 4.0 mm, and R4
antenna (referred to as antenna A) which was measured = 3.7 mm.
28 x 30 mm printed on material FR-4 substrate. The
circuit-board material has 1.6-mm thickness, relative
permittivity of 4.4, and loss tangent (tan δ) of 0.02.
The antenna is formed in the x-y plane with normal
direction parallel to the z-axis. The center strip and gap
of the CPW line are 3.6 and 0.3 mm, respectively, to
achieve a 50-Ω characteristic port impedance. The
improvement steps mentioned earlier contribute to
good impedance matching across a wide bandwidth
The antenna’s geometric parameters were optimized for
S11 ≤ −10 dB across the full frequency range using
ANSYS HFSS 11.0 electromagnetic (EM) simulation
software from Ansys. The final antenna geometry Fig. 3: Diagram Shows the Configuration of Proposed Antenna A
parameters were obtained as L1 = 28 mm, L2 = 30
mm, L3 = 13.5 mm, L4 = 5 mm, L5 = 1.5 mm, L6 =
REFERENCES
[1] J.R. James and P. S. Hall, Handbook of Microstrip Antennas.
London, U.K.: Peter Peregrinus, vol. 1, 1989.
[2] Kumar H, Upadhayay MD, Rai VK, Varsheney Leena, Kanth
RK, ― Multiband planar mircostrip antenna,‖ proceeding of the
IEEE International conference on antenna & propagation in
wireless communication, Sept. 16-20, 2011, vol. 2, 978-1-
4577-0048-4/11//$26.00 ©2011 IEEE, pp 1360-1363
[3] S.R. Best, ―On the Significance of Self-Similar Fractal
Geometry in Determining the Multiband Behavior of the
Sierpinski Gasket Antenna,‖ IEEE Antennas and Wireless
Propagation Letters, vol.1, no. 1, pp.22-25, 2002.
[4] C. Puente, J. Ponmeu, R. Pous and A. Cardama, ―On the
behavior of the Sierpinski Multiband Antenna,‖ IEEE
Fig. 4: Curves Compare the Simulated VSWRs of Antennas A and B Transactions on Antennas and Propagation, vol. 46, no. 4,
pp.517-524, April 1998.
[5] Junho Yeo and Raj Mittra, ―Design of Conformal Multiband
Antennas Based on Fractal Concepts‖, Microwave and
Optical Technology Letters, vol. 36, No. 5, 333-338, 2003.
[6] Tripti Luintel and Parveen F. Wahid, ―Modified Sierpinski
Fractal Antenna‖, IEEE/ACES International Conference on
Wireless Communications and Applied Computational
Electromagnetics, 3-7 April 2005, 578-581
[7] B. N. Biswas, Rowdra Ghatak, Rabindra K. Mishra, and Dipak
R. Poddar, ―Characterization of a Self-complementary
Sierpinski Gasket Microstrip Antenna‖, PIERS ONLINE, vol.
2, no. 6, pp. 698-701, 2006.
[8] Kanth, RK, W. Ahmad, S. Shakya, P, Liljeberg, L.-R. Zheng, H
Tenhu mous use of fractal in structure low cost, multiband and
Fig. 5: Compare the Simulated VSWRs of Antennas A, B, and C compact navigation antenna in proceeding of MMS, aug.2010,
Cyprus p 135.
[9] Kanth R.K., Qiansu Wan, Kumar,Harish, Pasi Liljeberg, Lirong
Zheng, Hannu Tenhunen,, Life Cycle Assessment of Printed
Antenna: Comparative Analysis and Environmental Impacts
Evaluation, Proceedings of 2011 International Symposium on
Sustainable Systems and Technology (ISSST)2011, PP 1,
ISBN:978-1-61284-394-0
[10] Kanth R.K., Harish Kumar Qiansu Wan, Pasi Liljeberg, Lirong
Zheng, Hannu Tenhunen,, Insight into Quantitative
Environmental Emission Analysis, of Printed Circuit Board,
Proceedings of International Conference on Environment and
Electrical Engineering (EEEIC) 2011, ISBN:978-1-4244-8779-
0
[11] Kanth, R.K.; Liljeberg, P.; Tenhunen, H.; Qiansu Wan; Amin,
Y.; Botao Shao; Qiang Chen; Lirong Zheng; Kumar, H.
Evaluating Sustainability, Environmental Assessment and
Toxic Emissions during Manufacturing Process of RFID Based
Systems, 2011 IEEE Ninth International Conference on
Dependable, Autonomic and Secure Computing (DASC),,1066-
Fig. 6: H-and E-plane Radiation Patterns were Made at (a) 3 GHz,
1071, IEEE, 2012
(b) 6 GHz, (c) 9 GHz, and (d) 12 GHz
[12] Kanth R.K.; Liljeberg, Pasi; Tenhunen, Hannu; Chen, Qiang;
Zheng, Lirong; Kumar, Harish, Comparative Toxic Emission
III. CONCLUTION Analysis in Production Process of Polymer and Paper Based
RFID Tags, Proceedings of International Conference on
Patch antennas successfully achieve sharp notches Environment and Electrical Engineering (EEEIC) 2012,
where needed across the frequency ranges of UWB ISBN:978-1-4577-1830-4 PP 184-187.
antennas inserting two nested C-shaped slots in the [13] Harish Kumar, Upadhyay M.D, ― Design UWB filter with
notched band for WLAN,‖ International journal of Antenna
radiating patch, notches of 3.3 to 3.6 GHz and 5.0 to & Propagation, ISSN: 1687-5869, E-ISSN: 1687-5877 Vol.
6.0 GHz representing WiMAX and WLAN 2012 article ID 971097, DOI 10.1155/2012/971097, page’s 1-4
frequencies, respectively can be obtained. Owing to the [14] Harish Kumar, Vibhor K Bhardwaj, Upadhayay MD, Analysis
simple structure and ease of impedance matching, the of Ultra wide band –band pass filter with WLAN notch,‖
ACEEE International journal on Communication (2012) DOI:
proposed antennas should be profitable for multiband 01IJCOM.3.1.503, Vol 03 Issue 01 July 2012 page(s) 45-47
applications
Abstract—In this paper we propose a novel compact adopted in conventional LPF, are not required in the
size microstrip filter using double dumbbell shaped funnel proposed DGS-LPF. The proposed DGS-LPF has only
type Defected Ground Structure (DGS). Here, we have
studied the response of Funnel Shaped dumbbell. It is largely broadened microstrip lines and DGS patterns on
observed that proposed DGS has a bandwidth of 5.23 the ground plane. It is expected that the broadened
GHz with large stopband as well as sharp transition width of microstrip line would provide an improved
between-3 dB to-35 dB. The quality factor is also very high power-handling capability.
good. The occupying area of the proposed DGS is less as
compared to others.
II. DESIGN & SIMULATION
Keywords: Low pass filter, DGS filter, funnel shaped
DGS For designing of sharp cut off transition, band stop
resonator[2], We take a microstrip line on FR4 epoxy
I. INTRODUCTION substrate of permittivity 4.4 and dielectric thickness of
h=0.8 mm and metal thickness t=0.035mm and width
A wireless communication system demands a large
of strip 1.2mm with a gap width of 0.5mm is calculated
number of base-station filters with not only an excellent
by closed form of expression . This design is simulated
in-band performance (i.e., low losses), but also a good
on High frequency 3-D simulator.
out-of-band spurious performance. Dielectric-resonator
filters are preferable for wireless base stations due to
their superior characteristics of a high quality factor
and miniaturization. However, cost reduction remains a
key limiting factor for the widespread use of dielectric-
resonator filters in base-station applications. There are
increased demands for low-loss dielectric-resonator
filters that are compact and capable of being
manufactured in a large quantity at a reasonable low
cost. A new seven-pole LPF design method using the
DGS has been proposed. The proposed DGS-LPF
neither has open stubs, nor high-impedance lines
because a very wide microstrip line has been adopted to Fig. 1: Funnel Shaped Dumbbell DGS Top view
realize the shunt capacitors. The method to calculate
the cutoff frequency of the LPF has been developed
based on the modeled equivalent inductance and
capacitance, which depends on the dimension of the
DGS pattern. In addition, the method to determine the
size of the DGS pattern, which exactly realizes the
required transformed inductance, has been proposed by
curve fitting with excellent accuracy. Furthermore, the
equivalent inductance and capacitance of the microstrip
line have been considered to adjust the size of the DGS
pattern to get an exact dimension of the DGS. Open
stubs, junction elements, very high-impedance lines,
Fig. 2: Funnel Shaped Dumbbell DGS Bottom view
and abrupt step junctions, which have always been
40 ISBN: 978-93-82880-20-2
Microstrip Low
L Pass Filterr using Double Dumbell
D Shaped
d Defected Grou
und Structure
the DGS has nott been used too improve thee stop band
charracteristics ofo the LPF. The researcchers have
commmented that for the equall area of slott head, any
shap pe of slot can be used. How wever, an equaal area only
ensu ures equal equivalent
e innductance an
nd not the
idenntical responsse of the DG GS circuit elements. A
particular shape of o the slot in a ground planee could also
be more
m appropriate to design a complete ciircuit on the
samme substrate. The T shape, siize, and orien ntation of a
slot can have an influence on performance of the filter
and other neighbooring circuits.
Here we repport a new D DGS based band
b accept
conffiguration in a microstriip line, by which we
Fig. 3: Freequency Responsse of Funnel Shapped Dumbbell DG
GS commbined the band b reject aand band acccept DGS
conffiguration witth microstripp to develop a compact
III. RESULTT & OBSERVATTION seveen pole low pass p Recently DGS have been
filter. R
We obbserve sharp cut
c off transittion [3]. Whenn we gainning interest for their plaanar form an nd ease of
vary the dimension
d of various param meter of defeected fabrrication. They are very usefful in the desiign of LPF,
ground strructure for opptimization, we
w realize thaat the BSF F and BPF, since a few DG GS can providde a cut-off
resonance frequency f 0 of funnel shaped Dumbbbell freqquencies and an a attenuationn pole withouut a need of
cell dependds on the physsical dimensioons of the celll. For periodic array ass compared w with EBG stru ucture. The
example, f 0 can be reduced
r by using
u smallerr gap latteer needs manyy units to build a filter. Onee advantage
of employing
e DG GS is that thee fabrication cost
c can be
between strip, larger area a of cell or
o larger disttance
between thhe two cells.. Since gap between stripps is redu uced with a siimultaneous im mprovement of the filter
generally limited by PCB fabriccation techniqques, perfformance. Annother advantaage is that th hey can be
increasing the size of thhe cell is the practical
p apprroach easily modeled usingu simple RLC circuits, especially
of reducinng the resonnance frequenncy. If theree are for microstrip
m linne structures. Because of thhe excellent
floating paatches inside thhe dumbbell-sshaped cell, itt will stopp-band and sloow-wave charaacteristics, thee DGS have
exhibit a shift
s toward lower values. To measuree the been n applied wiidely to micrrostrip circuiits such as
compactneess of the structures,
s w simulated the
we filters, amplifierss and antennaas. DGSs are realized by
transmissioon coefficientt of microstriip lines on toop of etchhing defects inn the backsidde metallic grround plane
these structures. The sim mulations are performed using a undeer a microstriip line. A basic and widely y used DGS
full-wave EM simulatoor. The substrate used inn the cell is composedd of two widde defected areas a and a
simulation has the samee parameters as a the FR4 withw a narrrow connectinng slot. A DGS-slot combin nation has a
board thickkness of 0.8 mm,m a dielectrric constant of ε r - simp ple structure and can be m modeled as a parallel LC
=4.4, and a loss tangentt of 0.0009. In I the simulattions, resoonator. Such a structure bloocks the signaal around its
the width of
o the center metal
m traces inn all the strucctures resoonant frequency and may be used to introduce i a
is kept at a fixed value and is the sam me as the widdth of widee stopband foor low-pass, bband-pass and d band stop
the microsttrip lines. filters. Such filters have sharpp transitions between
b the
The coompactness off the proposedd resonant celll can passsband and thee stopband, loow insertion loss in the
be understood based onn the parameteers extracted from passsband, wide stopband
s and high attenuaation in the
the equivaalent-circuit model
m proposeed by Woo et e al. stoppband.
We can compare thee extracted equivalent-ciircuit A new technnique is propoosed to reducee the size of
parameterss of the DGS S cell with coonventional seven
s the LPF. Using thhe resonant characteristics of the new
pole low pass
p filter. It is
i observed thhat the capacittance prop posed DGS, the t equivalentt circuit of th he proposed
of the prroposed cell is greater than t that off the DGS S unit sectionn is derived by using fieeld analysis
conventionnal low passs filter, while there is little meth hod. The new w DGS consiists of two Fu unnel areas
difference in the inducttances. Thus, we can concclude (equ uivalent to innductor), whicch are conneccted with a
that the extended
e gapp length resuults in increeased large slotted gapp (equivalentt to capacitorr), thus an
capacitancee, and the inccreased capaccitance leads to fo imprrovement in the t transition characteristicss of filter is
reduction and
a size reducction. achiieved. The following
fo obsservations aree measured
The quuasi-lumped element
e basedd on the photoonic- fromm the waveform m.
band gap (PBG) groundd plane and thee defected groound- f CU =8.34GH Hz
plane struuctures (DGS Ss) have beeen proposedd to f CL =5.23GHHz
improve reejection in the stop band of o LPF. How wever, Stop BW =3..1 GHz.
Bhagwant Institute
e of Technologyy, Muzaffarnaga h, India 41
ar, Uttar Pradesh
Proceedings of National Conference on Trends in Signal Processing & Communication (TSPC’13) 12th–14th April 2013
-40
-60
-80
0 2 4 6 8 10
freq, GHz
Abstract—A Broadband inverted E-H shaped probe technique with a thick air-filled substrate
microstrip patch anteena is proposed and experimentally provides the bandwidth enhancement,while the
investigated .The anteena employs E-H shaped patch with application of superstrate with inverted radiating patch
L-probe feed techniques. The designed anteena has a offers a gain enhancement,and the use of parallel (E
dimension of 80mm by 50mm.With the use of L-prob feed
and E-H microstrip patch.The antenna has achieved 30%
shaped) And series slot (H-shaped) reduce the size of
impedance bandwidth from 1.76GHz to 2.38GHz and a the patch.The H-ped slots also help to reduce the cross-
maximum gain of 9.37dBi. Stable radiation pattern across polarization level .e other hand world also provide the
the operating bandwidth are observed. In addition, a necessary protection for the patch from the
parametric study is conducted to facilitate the design & environmental
optimization process.
I. INTRODUCTION
Because of booming demand in wireless
communication system application,microstrip patch
antenna have attracted much interst due to their low
profil, ranging from narrow bandwidth and low
gain.However, light weight, ease of fabrication and
compatibility with printed circuits.However, they also
have some drawback, ranging from narrow bandwidth
to low gain[1].However, bandwidth enhancement and
size reducation are becoming major design
consideration for practical application of microstrip (a)
antennas due to the improvement of one of the
characteristics,which normally result in degradation of
the other.
This paper a new E-H shaped patch anteena for
enhancing the impedence bandwidth and gain.This
design employs contemporary techniques namely,the
L-probe feeding, inverted patch,& E-H shaped patch
techniques to meet the design requirement.By
extending the edge of E-shaped slot,a wider impedence (b)
bandwidth of 30% compared to the design reported Fig. 1: Configuration of the Proposed Patch Antenna
in[2-3] and gain of 9.37dbi compared to the design (a) Top View (b) Side view
reported in[4-5]
Effect.By incorporating amended E-shaped slot in
II. ANTENNA CONFIGURATION radiating edges, the bandwidth has been improved.In
addition, the antenna has wider impedence bandwidth
The configuration of the proposed inverted E-H in comparison with the antenna described in[2-3].The
patch antenna illustrated in figure 1.The proposed patch proposal radiating patch encompasses slots
with dimension of 80mm×50mm integrates both the E symmetrically adjacent to the excitation probe and
and H shaped slots on the same radiating element. The defining a capacitive load for compensating an
E-and H–shaped slots on the patch are shown in Figure inductance of the excitation probe antenna so as to
1(a).Table 1 shows the optimized design parameters obtain wideband operating frequency. Again single
obtained for the proposed patch antenna.The use of L-
ISBN: 978-93-82880-20-2 43
Proceedings of National Conference on Trends in Signal Processing & Communication (TSPC’13) 12th–14th April 2013
(b)
Fig. 3: Measured Radiation Pattern of the Antenna (a) Azimuth (b)
Elevation
REFERENCES
[1] Ramadan,A.,K.Y. Kabalan, A.EL-Hajj,S. Khoury, and M.AL-
Husseini “ A reconfigurable U-koch microstrip antenna for
wireless application,” Progress In Electromagnetic Research,
PIER 93,355–367,2009
[2] Wu, G-L., W.Mu, G.Zhao, and Y.-C. Jiao, “A noval design of
Fig. 4 (a): Measured Gain of the Antenna dual circulary polarized antenna feed by L-strip,” Progress In
Electromagnetic Research, PIER 79, 39–46, 2008.
[3] Sun, J. S. and S. Y. Huang,” A small 3-d multi-band antenna of
F shape for portable phones applications,” progress in
Electromagnetic Research Letters, VOL.9, 183–192, 2009.
[4] Abdulla, P. & A chakrabarty, “Rectangular waveguide feed
hemisphericai dielectric resonator antenna,” progress in
electromagnetic recharge PIER83, 225–244, 2008.
[5] Albooyeh, M., N. Kamjani, & M.Shobeyri, “A novel cross –slot
geometry to improve the impedance bandwidth of microstrip
antennas,” progress in Electromagnetic Research Letters,
VOL.4, 63–72, 2008.
[6] Islam, M. T., Shakib, and N. Misran, “Multi-slotted microstrip
patch antenna for wireless communication,” progress in
Electromagnetic Research Letters,VOL.10,11–18,2009
[7] Islam, M. T., Shakib, and N. Misran, “Design analysis of high
gain wideband L-prob feed microstrip patch antenna,” Progress
In Electromagnetic Research, PIER 95, 397–407, 2009.
Fig. 4 (b): Measured Efficiency of the Antenna [8] Foged, L.J. and A. Giacomini, “Wideband dual polarized
probes for near and farfield measurement systems,” IEEE
Antennas and propagation society international Symposium
IV. CONCLUSION 2008, 5-11, 2008.
[9] Ren, W., J.Y. Deng, and K.S Chen, “Compact PCB monopole
A.broadband microstrip antenna for gain and
antenna for UWB application,”Journal of Elestromagnetic
bandwidth enhancement is successfully developed. Wave and application, Vol. 21, No. 10, 1411–1420, 2007.
With the use of L-prob fed and E-H microstrip patch, [10] Denidni, D.A., N. Hassaine, and Q.Rao,” Broadband high-gain
the antenna has achieved 30% impedance bandwidth E-shaped microstrip antenna for high-speed wireless networks,”
Progress In Electromagnetic Research C, Vol. 1,105–111,
from 1.76GHz to2.38GHz and maximum gain of
2008.
9.37dBi. Techniques for microstrip broadbanding, size
reduction, high gain and stable radiation pattern are
carried out and experimentally verified in this design.
Abstract—This paper presents microstrip patch However, microstrip patch antennas suffer from a
antenna using aperture coupled fed. Various designs are number of disadvantages as compared to conventional
proposed by using partial substrate removal. The gain non printed antennas. Some of their major drawbacks
and radiation pattern of proposed antenna has been are the narrow bandwidth, low gain, and surface wave
investigated by using Ansoft HFSS simulator. The
antenna is designed for operating frequency 2.4 GHz. As
excitation that reduce radiation efficiency. However,
compare with conventional patch antenna, partial three types of losses i.e. conductor loss, dielectric loss
substrate is helpful to improve gain. In this paper three and surface wave loss results in lower gain of a patch
different structure is proposed and the designs is antenna. The conductor loss and the dielectric loss
confirmed through proper simulation results. depend on the quality of the materials being used such
as copper or gold, and the substrate, respectively. The
Keywords: Microstrip patch antenna, Aperture
coupling, return loss, radiation pattern. dielectric loss is dependent on the loss tangent of
materials and substrate thickness while the surface
I. INTRODUCTION wave loss depends on the permittivity of materials and
the substrate thickness [2]. Multiple techniques have
Microstrip antennas have a variety of applications been proposed, The simplest one is to use a thick
in widespread wireless communication system. In substrate that potentially could enhance the Appropriate
microwave and even millimeter-wave frequency ranges quality of conductor and substrate can reduce the
microstrip antenna has become an attractive tool, conductor and dielectric losses in Microstrip antenna,
because of the advantages like; it is very compact size, the gain of the patch antenna can further be enhanced
light and low cost [1]. It can be interconnected with by suppressing surface waves [3][4]. Another easy
other RF components neatly and require less space in a method is to replace the substrate of patch antennas
component configuration. It is also easy to be with air. But it becomes bulky, so another method is to
fabricated on the planar type substrate as antenna form cavity in substrate by partially removing the
component and also it is convenient to change the substrate. Which replace substrate with lower
shape of patch antenna for the purpose in limited space dielectric.
case by case. In this paper we propose new technique to enhance
Microstrip aperture coupled antenna is easy and gain by partial substrate removal. All Microstrip
cheap to manufacture, lightweight and planar. In this antennas are simulated on Ansoft “High frequency
scheme, a microstrip line on a back substrate excites a structure simulator (HFSS) “software.
slot in the ground plane, which then excites the patch
cavity. This scheme has the advantage of isolating the II. DESIGN OF APERTURE COUPLED
feeding network from the radiating patch element. It MICROSTRIP PATCH ANTENNA
also overcomes the limitation on substrate thickness
imposed by the feed inductance of a coaxial probe, so We are using conventional aperture coupled
that thicker substrates and hence higher bandwidths can microstrip patch antennas that will resonance at 2.4
be obtained GHz. Two substrate with dielectric constant Ɛr= 6.15
and loss tangent is of 0.002 of thickness h1=7.629mm
& h2=1.279 mm is stacked together. Ground plane is
formed between two substrates. The ground plane is of
125mm x 125mm. the centre of patch is about 70.1mm
from the substrate sides. Length & width of patch is
about L-16.8mm, W=25 mm respectively. A slot cut at
the center of patch and along the ground plane is taken
Fig. 1.1: Aperture Coupled Fed Microstrip Antenna. as Ls=19 mm & Ws=0.8 mm. the feed line is having
46 ISBN: 978-93-82880-20-2
Gain Enhancement in Microstrip Patch Antenna using Edge Modification of Patch and Partial Substrate Removal
Width Wf==1.85 mm. fig2.1 shows the 3d view of III. DESIGN OF APERTURE COUPLED MICROSTRIP PATCH
aperture coupled patch antenna and Fig 2.2 shows the ANTENNA WITH PARTIAL SUBSTRATE REMOVAL
Top and side view of Aperture coupled microstrip
The substrate S11 is partially removed around the
patch antenna
patch surface. This removal of substrate results in
reducing surface wave loss. This proposed substrate
removed antenna is going to achieve 10% gain
improvement as compare to conventional aperture
coupled microstrip antenna.
Design structure is shown in fig 3.1
The S11 plot & radiation pattern at 2.45 GHz is Fig. 3.2: Proposed Model Modified Aperture Coupled Microstrip
shown in fig 2.3(a) & (b) Patch Antenna by Partially Substrate Removal in 3D view
(Simulated in Ansoft HFSS)
V. CONCLUSION
The design and construction of Modified Substrate
& patch microstrip patch antenna using Aperture
coupled fed can be beneficial in term of advancement
in antenna’s technology that has very feasible value of
return loss. The combination of two substrate with
different surface area will be beneficial.
REFERENCES
[1] R. Garg, I. Bhartia, I. Bahl, and A. Ittipiboon, Microstrip
Antenna Design Handbook, Artech House, Boston, Mass, USA,
2001.
[2] R. Gonzalo, P. de Maagt, and M. Sorolla, “Enhanced patch-
antenna performance by suppressing surface waves using
photonic-bandgap substrates,” IEEE Trans. Microw. Theory
Tech., vol. 47, no. 11, pp. 2131–2138, Nov. 1999.
[3] Siew Bee Yeap, Zhi Ning Chen, “Microstrip Patch Antennas
With Enhanced Gain by Partial Substrate Removal,” IEEE
Trans. Microw. Theory Tech., vol. 58, no. 09, pp. 2811-2817,
Sep 2010.
[4] Siew Bee Yeap, Zhi Ning Chen, “Gain-Enhanced 60-GHz
LTCC Antenna Array With Open Air Cavities, IEEE Trans.
Microw. Theory Tech., vol. 59, no. 09, pp. 3470–3473, Sep
2011.
[5] H. Boutayeb and T. A. Denidni, “Gain enhancement of a
microstrip patch antenna using a cylindrical electromagnetic
crystal substrate,” IEEE Trans. Antennas Propag., vol. 55, no.
11, pp. 3140–3145, Nov. 2007.
[6] Manidipa Nath, “Wideband Micromachined Patch Antenna,”
International Journal of Scientific Research Engineering
&Technology (IJSRET) Volume 1, Issue2, pp 016-017, May.
2012.
[7] Awida, M.H., Suleiman, S.F., Fathy, A.E.: ‘Substrate-integrated
cavity backed patch arrays: a low-cost approach for bandwidth
enhancement’, IEEE Transactions on antennas and propagation,
2011, 59, pp. 1155–1163.
[8] D.M. Pozar, “Microstrip antennas,” Proc. IEEE, vol. 80, pp.
79–91, Jan. 1992.
Fig. 4.1: (a) Top View of Proposed Design with Modified Patch
(b) Aperture Coupled Microstrip with Modified Patch Antenna by
Partially Substrate Removal in 3D view (Simulated in Ansoft HFSS)
(c) Top View of Simulated Antenna in HFSS
I. INTRODUCTION
Modern communication systems require antennas
with more bandwidth and smaller dimensions
compared to conventional antennas. Fractal geometry is
a very good solution to fabricate these antennas.
Fractals are created from self-similar elements which
are iterated in various directions and increasing
iterations does not change their entire form because Fig. 1.2: First Iteration
their small sections are reduced-size copy of the whole.
Self-similarity of fractals causes multi-band and broad-
band properties of antennas [1-4].
Fractals are highly convoluted and jagged shapes
in which these discontinuities increase bandwidth and
effective radiation of antennas. Fractal antennas can
place long electrical length into small volume using
their ability of space-filling [5-7]. In this paper another
shape of fractal is presented.
Dimensions of this fractal antenna are 45mm circle
as generator and its bandwidth is 3GHz Fig. 1.3: Second Iteration
ISBN: 978-93-82880-20-2 49
Proceedings of National Conference on Trends in Signal Processing & Communication (TSPC’13) 12th–14th April 2013
IV. CONCLUSION
The concepts of fractals is applied to analyze
proposed antenna as per the result proposed design is
most suited for dual band operation. The self similar
nature of the geometry of the fractal antenna results in
very similar antenna characteristics across many band
as defined by the geometry.
Fig. 4.3: VSWR for Third Iteration
Abstract—A new antenna structure comprising a performance has been validated by the measured return
semicircular micro strip patch alongside a small loss and radiation patterns.
rectangular shape ground proximity fed by a micro strip
line is studied. On a thin substrate this antenna achieves
in the range of 5.8–12.9 GHz an impedance bandwidth of
almost 75%. Details of the antenna design, simulation and
measured results on the return loss and the E and H-
plane radiation pattern of the studied antenna are
presented.
I. INTRODUCTION
The low profile, light weight and low cost of
manufacturing of micro strip patch antennas have made
them attractive for many applications. The modern
trends in communication systems require wide
bandwidth and small size, low profile antennas. Micro
strip patch antennas on a thin dielectric substrate
inherently have the disadvantage of narrow impedance Fig. 1: Geometry of the Antenna
bandwidth. To increase the bandwidth of a single layer
micro strip patch antenna several configurations have II. ANTENNA GEOMETRY
been proposed by researchers such as placing parasitic
patches on the same layer with the main patch [1], chip The geometry of the antenna is shown in Fig. 1,
resistor loading [2], E-shaped patch [3], placing a U- where a semicircular patch with radius r = 6 mm is
slot on the patch [4], planer micro strip fed tap placed co-planar to a finite ground plane that has a
monopole antenna [5], rectangular slot antenna with rectangular shape with size of G W = 30 mm ×
patch stub [6], Vivaldi antenna [7] and square ring GL = 10 mm. The dielectric substrate used is FR4
patch with shorting posts [8]. Majority of these works of dimension 30 × 30 mm2 with thickness h = 1.6
carried out are based on relatively thick substrates. mm. The separation distance between the patch
Indeed, there are many different methods proposed in and the ground is d = 2 mm. The patch is
the literature to tackle the narrow-band width problem, proximity fed by a 50C micro strip line with line length
[9–11] none of which uses thin substrate to achieve and width FL = 10 mm and FW = 3 mm,
wide bandwidth performance. respectively.
In this paper, a low profile micro strip patch
antenna is studied. Unlike the usual method of
placing the patch antenna on top of a ground plane,
the patch is placed alongside a small rectangular
ground co planar to it. The thickness of the
substrate used is approximately 0.02 λ (1.6mm).
The size of the ground and the radius of the patch
and their relative position are optimized to achieve a
wide bandwidth. The optimized antenna structure
achieves in the frequency range of 5.86–12.86
GHz an impedance bandwidth of almost 75%. E
and H-plane radiation patterns shows slight tilt in
the main beam position but the patterns are almost
unchanged in the operating band. The achieved Fig. 2: Top and Bottom Views of the Antenna
ISBN: 978-93-82880-20-2 53
Proceedings of National Conference on Trends in Signal Processing & Communication (TSPC’13) 12th–14th April 2013
IV. CONCLUSION [5] Eldek, A. A., “Numerical analysis of a small ultra wideband
micro strip-fed tap monopole antenna,” Progress In
In this paper a co-planar micro strip patch antenna Electromagnetics Research, PIER 65, 59–69, 2006.
with wide bandwidth behavior has been studied. [6] Eldek, A. A., A. Z. Elsherbeni, and C. E. Smith, “Rectangular
slot antenna with patch stub for ultra wideband applications and
Simulated as well as measured results are presented phased array systems,” Progress In Electromagnetics Research,
for a semicircular shape patch antenna. Compared PIER 53, 227–237, 2005.
with other micro strip patch antennas of high [7] Mehdipour, A., K. Mohammadpour-Aghdam, and R. Faraji-
bandwidths this structure has the attractive features of Dana, “Complete dispersion analysis of vivaldi antenna for
ultra wideband applications,” Progress In Electromagnetics
low profile, smaller patch size and being simple to Research, PIER 77, 85–96, 2007.
design. Optimization of the structure gives 73% [8] Row, J. S. and S. H. Chen, “Wideband monopolar square ring
impedance bandwidth with reasonable bidirectional patch antenna,” IEEE Transactions on Antennas and
patterns suitable for many applications. Propagation, Vol. 54, No. 4, 1335–1339, April 2006.
[9] Joardar, S. and A. B. Bhattacharya, “Two new ultra wideband
dual polarized antenna-feeds using planar log periodic antenna
REFERENCES and innovative frequency independent reflectors,” Journal of
Electromagnetic Waves and Applications, Vol. 20, No. 11,
[1] Kumar, G. and K. C. Gupta, “Directly coupled multiple 1465– 1479, 2006.
resonator wide-band micro strip antenna,” IEEE Transactions [10] Chen, X. and K. Huang, “Wideband properties of fractal bowtie
on Antennas and Propagation, Vol. 33, 588–593, June 1985. dipoles,” Journal of Electromagnetic Waves and Applications,
[2] Wong, K. L. and Y. F. Lin, “Small broadband rectangular micro Vol. 20, No. 11, 1511–1518, 2006.
strip antenna with chip-resistor loading,” Electron. Lett., Vol. [11] Chair, R., A. A. Kishk, K. F. Lee, C. E. Smith, and D. Kajfez,
33, 1593–1594, 1997. “Micro strip line and CPW FED ultra wideband slot
[3] Yang, F., X.-X. Zhang, X. Ye, and Y. Rahmat-Samii, “Wide- antennas with U-shaped tuning stub and reflector,”
band E-shaped patch antennas for wireless communications,” Progress In Electromagnetics Research, PIER 56, 163–182,
IEEE Transactions on Antennas and Propagation, Vol. 49, No. 2006.
7, 1094–1100, July 2001. [12] Sze, J.-Y. and K.-L. Wong, “Bandwidth enhancement of a
[4] Weigand, S., G. H. Huff, K. H. Pan, and J. T. Bernard, micro strip line fed printed wide-slot antenna,” IEEE
“Analysis and design of broadband single layer U-slot micro Transactions on Antennas and Propagation, Vol. 49, 1020–
strip patch.antennas,” IEEE Transactions on Antennas and 1024, July 2001
Propagation, Vol. 51, No. 3, 457–468, March 2003.
Abstract—This paper presents the design of a octagon incorporation of fractal technology in micro strip
shape fractal antenna. This octagon shape fractal antenna antennas. This paper presents the novel antenna design
has been designed on substrate with dielectric constant 4.4 based on self-similarity of fractal geometry to achieve
and substrate thickness 1.6 mm. The antenna has been the multiband, wideband as well as size reduction of
simulated using IE3D software which work on principle of
MOM. The simulated result shows the resonance at
the antenna.
1.24GHz, 1.87GHz, 1.8GHz, 2.13GHz, 2.5GHZ and
2.9GHz for different iteration. The measured pattern of II. DESIGN APPROACH
antenna displays good mono-polar mode radiation and the
The octagon shaped fractal antenna has been
gain is found to be stable in most of the transmission band.
This antenna is very suitable for Wideband (WB) stacked on square shape of 32x32mm2 dimension.
application. Substrate for the design has dielectric constant 4.4 and
thickness 1.6mm. Octagon shape antenna of radius
Keywords: Antenna, Resonant Frequency, Fractal 40mm is stacked on square shape micro strip line as
Geometry, Multi-band
shown in Fig 1a. From this conventional structure
octagon shaped geometry is generated to achieve first
I. INTRODUCTION iteration as in Fig 1b.Similarly for other two iterations
There has been an ever-growing demand, in as in fig1c and fig1d.
military as well as the commercial applications for
antenna design that possess the small size, ease of
fabrication, conformal, low cost and multiband
capabilities. Micro strip antenna [I] is one of them to
fulfill these requirements. The limitations of the micro
strip antenna are its narrow bandwidth, efficiency, and
size. The micro strip antenna is efficient radiator around
half wavelength. There are various approaches to
reduce the size of the antenna without much affecting
the antenna performance. The application of the fractal
geometry is one of the techniques to reduce the size of
the antenna as well as to provide multiband and
wideband properties. An excellent overview of various
Fig. 1(a): Base Shape
useful techniques for designing compact (i.e.,
miniature) antennas may be found in [1-3]. Moreover, a
number of approaches for designing multi-band
antennas have been summarized in [4].Recently, the
possibility of developing antenna designs that exploit in
some way the properties of fractals to achieve these
goals, at least in part, has attracted a lot of attention.
The term fractal which means broken or irregular
fragments was originally coined by Mandelbrot to
describe a family of complex shapes that possesses
inherent self-similarity or self-affinity in their
geometrical structure [3-5]. The recent developments in
the wireless [5] technology functional at multiple
frequency demands multiband [6] or multifunctional
antennas with broader bandwidth. It could be possible
Fig. 1(b): First Iteration
to achieve these objectives with the help of
ISBN: 978-93-82880-20-2 57
Proceedings of National Conference on Trends in Signal Processing & Communication (TSPC’13) 12th–14th April 2013
Fig. 2(a): Variation in Return Loss for First Iteration Fig. 3(a): First Iteration
IV. CONCLUSION
The octagon multiband fractal antenna resonating
at multiple frequencies has been designed, It is
Fig. 3(c): Third Iteration
concluded that incorporation of fractal geometry in
For the designed antenna Fig.4.a, Fig.4.b and conventional antenna reduces the size of antenna by
Fig.4.c shows the smith chart for different iteration. shifting the resonant frequency.
REFERENCES
[1] [K. Fnjimoto, A. Henderson, K. Hirasawa, and J.R. James,
Small Antennas. New York, John Wiley & Sons, Research
Studies Press, 1987.
[2] A.K. Skrivervik, J.F. Zurcher, o. Staub, and J.R. Mosig, "PCS
Antenna Design: The Challenge of Miniaturization," IEEE
Antennas and Propagation Magazine, 43,4, August 2001,
pp. 12–26.
[3] Jaggard, D.L.,"On Fractal Electrodynamics," Recent Advances
in Electromagnetic Theory, H. N. Kritikos and D. L. Jaggard
(eds.), I 83-224,Spring- Verlag, Yew York, 1990.
[4] Werner, D. H. and Ganguly S., "An Overview of Fractal
Antenna Engineering Research", IEEE, Antenna and
Propagation Magazine, Vol. 45, No.1, 38–57, 2003.
[5] N. Cohen, "Fractal Antenna Applications in Wireless
Telecomm." in Proc. Professional Program Elect. Industry
Forum 1997, pp. 43–49.
[6] Peunte, C., Romeu 1. Pous R., and Cardama A.,"On the
Behaviour of the Sierpinski Multiband Antenna ", IEEE Trans
on AP-46, No.4, 517–524, 1998.
Fig. 4(a): First Iteration
Abstract—Fractal geometry involves a recursive smaller physical area. The increased electrical length
generating methodology that results in contours with leads to a lower resonant frequency, which effectively
intricate fine structures. This geometry, has been used to miniaturizes the antenna. The second benefit is that the
model complex objects found in nature such as clouds and increased electrical length can raise the input resistance
coastlines, has property of space-filling that can be utilized
of the antenna when it is used in a frequency range as a
to miniaturize antenna. These contours are able to add
more electrical length in less volume. Due to self-similarity small antenna.
property antenna executes high directivity ,improved
bandwidth and multiband behavior. This paper present a II. ANTENNA SPECIFICATION
novel conformal multi band antenna design, based on
fractal concepts, for application in 1-6GHz band. This
This fractal antenna is designed from normal
paper is based on square-circle shape island fractal type square patch of size 35mm x 35mm. The generator of
configuration, printed on a dielectric substrate backed by the shape is shown in figure 1.a. Circle of radius 35mm
a ground plane, and propose novel approach to enhance created inside the square geometry.
its multiband performance in terms of impedance
matching characteristics as well as radiation pattern at
different operating frequencies. The method of moments is
used to analyze the antenna structure.
Keywords: Fractal, micro strip antennas, method of
moments
I. INTRODUCTION
Fractals represent a class of geometry with very
unique properties that can be enticing for antenna
designers. Fractals are space-filling contours, meaning
electrically large features can be efficiently packed into
small areas. Since the electrical lengths play such an
important role in antenna design, this efficient packing
can be used as a viable miniaturization technique [1-3].
In this article, we will summarize some of the unique
features of fractal geometries, and draw conclusions Fig. 1(a): Proposed Generator
about the utilization of these complicated intricate
structures in manufacturability of antenna. It will be Figure 1.b. shows First iteration in which the shape
shown how to quantify the space-filling ability of is generated within the generator structure. By again
fractal geometries, and how this correlates with repeating the concept we get the second iteration shown
miniaturization. Fractal elements can be utilized in in figure1.c.
arrays to tightly pack elements and to reduce mutual Results show that the location of feeding has many
coupling, resulting in improved, wider scan angles. The effects on antenna parameters and by changing it,
space-filling abilities of fractals can exhibit two benefits antenna parameters such as bandwidth and gain
over the conventional antenna. The first benefit is that changes.
the increased space-filling ability of the fractal antennas
means that more electrical length can be fitted into a
60 ISBN: 978-93-82880-20-2
Novel Design of Improved Bandwidth Fractal Antenna
IV. CONCLUSION
The concepts of fractals can be applied to improve
the bandwidth of smaller size antennas. Also, the
proposed structure has small dimensions of 35×35 mm².
Due to the simulation results, it is conclude that it is a
improved antenna for 1-6 GHz range.
REFERENCES
[1] J.P. Gianvittorio, “Fractal, MEMS, and FSS Electromagnetic
Devices Miniaturization and Multiple Resonances” Dissertation,
University of California, Los Angeles 2003.
[2] N. Cohen, “Fractal Antenna Application in Wireless
Telecommunications” Proceedings of Electronics Industries
Forum of New England, 1997, pp. 43–49.
[3] K.J. Vinoy, “Fractal Shaped Antenna Elements for Wide and
Multi–Band Wireless Applications” Thesis, Pennsylvania, Aug.
Fig. 4(b): Radiation Pattern at 1.1 GHz 2002.
[4] J.P. Gianvittorio and Yahya Rahmat Samii,”Fractal Antennas: A [7] C. Puente, J. Romeu, R. Pous, J. Ramis and A. Hijazo, ”Small
Novel antenna Miniaturization Technique and Applications” but Long Koch Fractal Monopole” Electronics Letters,1998.
IEEE Antenna and Propagation Magazine, Vol. 44 No.1, [8] John P. Gianvittorio and Yahya Rahmat-Samii “A Novel
Feb 2002. Antenna Miniaturization Technique, and applications” IEEE
[5] J. Gouyet,”Physics and Fractal Structures” New York: Antenna & Propagation Magazine Vol.44, Feburary 2002.
Springer, 1996.
[6] D.H. Werner, and Raj Mittra, “Frontiers in Electromagnetic”,
New York: IEEE Press 2000.
Abstract—In this paper a multiband and wideband There are three resonant frequencies f1,f2,f3.
Fractal antenna is proposed. The antenna geometry is Where f1=1.08 GHz
based on a perturbed planar Sierpinski fractal shape. This f2=3.45GHz
antenna exhibit a good impedence matching and very high f3=5.22GHz
efficiency at three frequency band 1.08 GHz, 3.45 GHz
and 5.226 GHz. In simulation of this antenna a software
The upper thresholds of the antenna length and
that works on the principle of Method of Moments is used. width have been set to a ≤80 mm and b≤80 mm.
Starting from these project specifications, the
I. INTRODUCTION synthesized values of the antenna descriptors are
reported in Table I. This parametrization has been
With the advance of wireless communication determined by exploiting the PSO-based optimization
systems and increasing importance of other wireless approach proposed in [21]
applications, wideband and low profile antennas are in The antenna satisfies the geometrical constraints
great demand for both commercial and military having a maximum dimension of 76 mm corresponding
applications. Multi-band and wideband antennas are to an electrical dimension of 0.19λ1close to that of a
desirable in personal communication systems, small quarter-wave resonant monopole printed on the same
satellite communication terminals, and other wireless dielectric substrate (i.e 0.19λ1., ) [22], while providing a
applications. Wideband antennas also find applications dual-band behavior instead of the classical single band
in Unmanned Aerial Vehicles (UAVs), Counter enabled by the standard monopole.
Camouflage, Concealment and Deception (CC&D), Finally, it is worth pointing out that, dealing with a
Synthetic Aperture Radar (SAR), and Ground Moving monopole antenna, the optimized antenna has to be
Target Indicators (GMTI). Some of these applications mounted over a perpendicular ground plane in order to
also require that an antenna be embedded into the analyze its performances from both the numerical and
airframe structure. Traditionally a wideband antenna in the experimental point of view. For the integration into
the low frequency wireless bands can only be achieved communication devices, a possible solution, similarly to
with heavily loaded wire antennas, which usually means [23], exploits the fractal patch to excite a slot antenna
different antennas are needed for different frequency with ground plane coplanar to the patch itself.
bands. Recent progress in the study of fractal
antennas[1]-[7] suggests some attractive solutions for
using a single small antenna operating in several
frequency bands.
64 ISBN: 978-93-82880-20-2
Design of Multiband and Wideband Sierpinski Fractal Antenna
TABLE 1
Geometry in mm
a b c d e F g h
76 33 16.5 17.989 20.89 5 0.5 38.625
A. Experimental Results
All the simulated data have been obtained by
means of an electromagnetic simulator based on the
method of moments (MoM) [24]. Fig. 4
Dimension of ground plane is 33 ×76 mm2.It is
clear that from the simulated graph that it has S11 IV. CONCLUSION
parameter is ≤ - 10 dB for three frequency band In this letter a multiband and wideband antenna is
proposed. It has three frequency band 1080 MHz, 3450
III. SIMULATED RESULT MHz, and 5226 MHz. Due to advancement in cellular
S11 Parameter technology and LTE of cellular technology this can be
very useful ,depend on application
REFERENCES
[1] R. Azaro, E. Zeni, P. Rocca, and A. Massa, “Synthesis of a
Galileo and Wi-Max three band fractal-eroded patch antenna,”
IEEE AntennasWireless Propag. Lett., vol. 6, pp. 510–514,
2007.
[2] E. Zeni, R.Azaro, P. Rocca, andA.Massa, “Aquad-band patch
antenna for GalileoandWi-Max services,” Electron. Lett., vol.
43, pp. 960–962, 2007.
[3] R. Azaro, F. Viani, L. Lizzi, E. Zeni, and A. Massa, “A
monopolar quad-band antenna based on a Hilbert self-affine pre-
fractal geometry,” IEEE Antennas Wireless Propag.Lett., vol. 8,
pp. 177–180, 2008.
Fig. 2 [4] R. Azaro, E. Zeni, P. Rocca, and A. Massa, “Innovative design
of a planar fractal-shaped GPS/GSM/Wi-Fi antenna,” Microw.
Opt.Technol. Lett., vol. 50, pp. 825–829, 2008.
[5] L. Lizzi and G. Oliveri, “Hybrid design of a fractal-shaped
GSM/UMTS antenna,” J. Electromagn.Waves Appl., vol. 24, pp.
707–719, 2010.
[6] J. Anguera, C. Puente, C. Borja, and J. Soler, “Fractal-shaped
antennas: A review,” in Encyclopedia of RF and Microwave
Engineering, K. Chang, Ed. Hoboken, NJ: Wiley, 2005, vol. 2,
pp. 1620–1635.
[7] C. T. P. Song, P. S. Hall, H. Ghafouri-Shiraz, and D.Wake,
“Sierpinski monopole antenna with controlled band spacing and
input impedance,” Electron. Lett., vol. 35, pp. 1036–1037, 1999.
Fig. 3 [8] Dual-Band Printed Fractal Monopole Antennafor LTE
Applications, Leonardo Lizzi, Student Member, IEEE, and
Radiation Pattern-Antenna and Radiation Andrea Massa, Member, IEEE IEEE Antennas and Wireless
Efficiency. Propagation Letters, Vol. 10, 2011.
Abstract—The design of active GPS antennas involves volume and thin profile characteristics. The size of
essentially three components: a patch element, filter, and patch antenna is basically determined by its resonance
low noise amplifier (LNA). Each of these components
forms an important part of the overall antenna assembly length and width. The reduction of the patch size can
and is critical to performance. Microstrip patch be achieved by using patch substrate material w ith
technology is often used to create the antenna element for very high permittivity and small substrate height
automotive GPS applications. In this paper, the effect of [2]. But, in this case, the low radiation efficiency will
shorting wall length on the circular polarization of shorted reduce the antenna gain. Various techniques have
microstrip patch antenna is examined. The theoretical
analysis is based on the High Frequency Software been explored to develop small size antenna. One of
Simulator (HFSS). A designed antenna for the GPS these techniques is achieved by using shorting wall [3].
application at the civilian GPS frequency (1227.60 MHz) For a rectangular MPA, its resonance frequency is on
has been fabricated and measured. The measured data for the order of half-wave length. The length of the antenna
the fabricated antenna is taken by 8719 ES vector network
is reduced by a factor of two if the half wave patch is
analyzer. Results of the designed antenna show good
circularly polarized with a reduction in its size by 25 % short-circuited to the ground at the zero potential
from the conventional rectangular microstrip patch planes. The result is a resonance rectangular patch with
antenna length of order λ/4, called a quarter-wave patch or
Keywords: Circular Polarization Antenna, Shorted Shorted Rectangular Microstrip Patch Antenna. The
Microstrip Patch Antenna, Global Positioning System objective of this paper is to study the effect of shorting
Antenna. wall length on the polarization of the SRMPA and to
design a compact circularly polarized patchantnna
I. INTRODUCTION operates at the civilian resonance frequency (1.227
In the last few years the Global Positioning GHz) for the Global Positioning System (GPS).
System (GPS) has been used in a variety of
applications for which new and more restrictive II. ANTENNA CONFIGURATION AND
requirements in the design of the receiving antenna SIMULATED RESULTS
have been introduced. In particular, for high- precision The geometry of the antenna is shown in Figure
GPS applications, such as differential GPS, GPS- based 1.The feed point is centered on the x-axis and the
spacecraft attitude determination or geodetic surveying, shorting wall is at the patch edge. The antenna is
a receiving antenna with superior rejection to multipath circularly polarized (CP) by the single feed technique
signals is required. with a truncated segment setting to the edge of the
Many applications in communications and radars patch with equal side length ∆L and area of δS/2.
require circular or dual linear polarization, and the The SRMPA has a side length L/2 and width W
flexibility afforded by microstrip antenna technology [4]. The input impedance and resonance frequency of
has led to a wide variety of designs and techniques to the SRMPA can be controlled by changing the length
fill this need [1]. In recent years, the demand for of the shorting wall [5]. To study the effects of shorting
compact mobile telephone handsets has grown. wall length on the SRMPA parameters, a proposed
Handsets with size of a pocket have begun appearing in antenna is specified as in Table 1 and simulated with
the market and, as the demand for increased electronic HFSS software with different shorting wall lengths.
mobility grow. A small antenna size is required as one HFSS is an interactive software package, developed by
of the important factors in portable mobile Agilent technologies. It is a full-wave field solution for
communication systems. The Microstrip Patch any arbitrarily 3-D passive structures.
Antennas is widely being used because of its low
66 ISBN: 978-93-82880-20-2
An
nalysis and Desiign of a Microsttrip Patch Anten
nna with Circular Polarization
the variation
v of thhe shorting waall length (SW
W) with the
trunccated length (∆L)
( for optim
mum axial ratiio.
New SW Length
L New Frequency New ∆L (mm) for III. COMPA
ACT CIRCULARRLY POLARIZE
ED
(mmm) (GHz) Best AR PARTIALLY S
SRMPA
96.322 1.227 4.2
95.322 1.222 4 o the previouus study, the design of a
As a result of
94.211 1.221 3.6 rightt hand circularly polariized (RHCP P) partially
93.600 1.218 3.2 SRMMPA operatingg at the civilian resonancee frequency
80.322 1.215 2.9
of th
he Global Poositioning Sysstem is consiidered. The
75.411 1.211 2.6
68.377 1.175 2.4 antennna configuraation is shownn in Figure 1 and
a antenna
64.322 1.155 2.2 parameters calculaated from HFS SS are given in
i Table 3.
TABLE 3: SPEECIFICATIONS OF TTHE RHCP PARTIIALLY
SMPA FOR TH HE GPS
fr εr h W L/2 ∆L
1.2
227GHz 2.2 1.58mm 6..2cm 3.2cm 2.2.0mm
TABLE
A 4: SIMULATEED RESULTS OF TH
HE RHCP PARTIA
ALLY SMPA
FOR THE G
GPS
Parameeters Partially SRMP
Beam m width 72 0
Direectivity 7.4 dB
Axiaal ratio 1.00
Retuurn loss -27.12
Bhagwant Institute
e of Technologyy, Muzaffarnaga h, India 67
ar, Uttar Pradesh
Proceedings of National Conference on Trends in Signal Processing & Communication (TSPC’13) 12th–14th April 2013
simulated results are given by HFSS. Fig. 4 shows the The simulated radiation pattern of the RHCP
comparison between the simulated and measured return partially SRMPA at 1.227 GHz is shown in Fig. 5. The
loss for the proposed GPS antenna. However, there is a simulated axial ratio results of the RHCP partially
little small deviation between the measured and SRMPA is shown in Fig. 6. The measured and
theoretical results. This deviation may be due to simulated results are accepted and satisfy the
fabrication tolerance. In addition, the tangent loss and requirements for a GPS.
bonding of the coaxial feeding are not considered in the
simulator software calculation. IV. CONCLUSION
The effect of short wall length on the antenna
polarization is examined. Simulations and
measurements on the partially shorted rectangular
microstrip patch antenna have provided a useful design
for a compact circularly polarized antenna operating at
the frequency of 1.227 GHz for the GPS applications. A
reduction in antenna size by 24.6 % from the
conventional antenna is achieved.
ACKNOWLEDGMENT
The authors would like to express their cordial
thanks to Jaipur National University, Jaipur, and BIT
Mesra for providing the facility of Research in their
University. Thanks to Mr. T.P. Mehta and late Mrs.
Fig. 5: Simulated Radiation Pattern Results of the RHCP Shanti Devi Mehta for their cooperation and
Partially SRMPA encouragement.
RFFERENCES
[1] Kin-Lu Wong, Compact and Broadband Microstrip Antennas,
John Wiley & Sons, nc.2002
[2] J.R. James, P. S. Hall, and C. Wood, Microstrip Antenna Theory
and Design, Peter Perigrinus, London, 1981
[3] Sanad, M. "A Small Size Microstrip Antenna Having a Partial
Short Circuit" IEEE International Conference on Antenna and
Propagation, vol. 1, pp. 465-471, April1995
[4] J.R. James and P. S. Hall, Handbook of Microstrip Antennas,
Peter Perigrinus Ltd., London, 1989
[5] T. Taga and K, Tsuna Kawa “Performance Analysis of a
Built-Planar Inverted-F Antenna for 800 MHz Band Portable
Radio Units” IEEE Journal on selected areas in communication,
Vol.5, NO.5, PP.921- 929, June 1987
TRACK II
EMERGING TECHNOLOGY TRENDS
ISBN: 978-93-82880-20-2 3
Proceedings of National Conference on Trends in Signal Processing & Communication (TSPC’13), 12th–14th April 2013
ISBN: 978-93-82880-20-2 71
Proceedings of National Conference on Trends in Signal Processing & Communication (TSPC’13) 12th–14th April 2013
Abstract—Technology [use is] becoming second database attracts the criminals andleads to a sharp
nature to the whole generation’ increase of computer crime attacks on SNS. Criminals
This paper demonstrate the concept, how various can perform different kinds of attacks to obtainsensitive
dangers of technology affected to the human in the form of personal information. In addition to the
crime. In Our Gadget-Obsessed society, the daily use of privacyprotection problem, the financial loss and the
technology has become second nature. Seeing people damage on Reputation of one person or organization is
listening to music, texting, or updating their Face book
status while walking down the street is commonplace. The also significantlylarge.In order to raise the public
benefits of living in a technologically advanced world are awareness and to provide a solidunderstanding and
abundant, but there are also dangers associated with summary of SNS, we have studied therelated criminal
everyone’s favorite electronic favorite electronic toys and activities on SNSs, the possible safetymeasurements
social networking sites. and the legislation works from the start to therecent
In recent years, there have been multiple reported
development of SNSs.
instances of online victimization. Social networking sites
have been held partially responsible for many different In order to raise the public awareness and to
scams, stalking and harassment, and murder, yet many provide a solidunderstanding and summary of SNS, we
social media networkers continue to post personal have studied therelated criminal activities on SNSs, the
information online. possible safetymeasurements and the legislation works
Threats online, through different mediums like from the start to therecent development of SNSs.
Facebook, Twitter, or just general email, are one of the
Social network sites have made communication for
most common problems we get in schools,’ he said. ‘How
many [people] have more than 100 friends [on Facebook, peoplenot only convenient, but also instantaneous
Twitter, etc]? It’s impossible to know all 200 people very connections acrossthe boundaries. According to the
well, so the best way to protect yourself is to be careful as report of Criminal Use ofSocial Media [2], before 2012,
to who you speak to and who you befriend on those social there were 750 million activeusers in Facebook, 500
networking mediums.’ million users in YouTube, 200 millionusers in Twitter
Social networking sites have become an integral part
and 100million users in LinkedIn.
of the lives of most of the youth today. It has encouraged
new ways to communicate and share information. The Table 1 Shows how frequency people use SNS.
relative freedom afforded by social networking sites has TABLE 1: THE STATISTICS OF THE USAGE OF SNSS
caused concern regarding the potential of its misuse by
individual patrons. SNSs Users (Million) Statistics
Facebook 1120 700 billion minutes per month spent
This questions the safety and security of the users,
YouTube 500 92 billion views per month
especially girls who are more preyed upon particularly in
Twitter 256 65 million tweets are created per day
relation to online sexual predators. These sorts of cyber LinkedIn 589 Users are across 1 million companies
crimes leave an everlasting scar in the minds of the users
and change the way in which they communicate in the Social networking sites, email, instant messaging,
social networking sites and sometimes even force them to video- and photo- sharing sites and comment posting
completely opt out of the social networking sites. This
study deals with how girls are affected by cyber crimes in
are all tools that help people to communicate and
social networking sites and how those crimes impact their socialize with each other (Mooney, 2009). The first
social networking pattern. social networking site SixDegrees.com was launched in
1997. It allowed users to create profiles, list their
Keywords: Cyber Crimes, Social Networking, Users,
Safety, SNS. friends and surf the friends list (Boyd, & Ellison, 2007).
Starting from 1997 to 2010 there are some 1.5 billion
I. INTRODUCTION users of social networking websites (Kreutz, 2009).
People join social networking sites because it gives
Social Network Service (SNS) [1], e.g. Facebook, them an opportunity to express their views, a feel of
Twitter,and LinkedIn etc. changes the world. Each of independency and self esteem. It’s just a click way to
these ownsthousands of millions of users’ personal create a virtual profile and connect with millions of
information. Owningsuch a large user information users across the world, becomes more interesting
ISBN: 978-93-82880-20-2 75
Proceedings of National Conference on Trends in Signal Processing & Communication (TSPC’13) 12th–14th April 2013
because you can stay connected with old friends and profile one can easily determine likely physical location
family members. of a person. Social networking sites are becoming a
According to ComScore, a leader in measuring the threat to the privacy of an individual.
digital world, 84 percent of India’s total internet
visitors are users of SNS. India is the seventh largest A. Impact of Unknown Friends
market worldwide for social networking after the U.S., When you accept friend request from strangers,
China, Germany, Russian Federation, Brazil and the you put yourself at severe risk. The men at other side
U.K. Facebook captures the top slot among SNS in start exchanging words very cordially and within some
India with 20.9 million visitors. Interestingly, Indian days you have a faith on them. Then they start the mind
social networking audience has increased by 43 percent game and start gathering more personal information. If
within 2009–10 (‘Face book captures top,’ 2010). they get to know your email address, they could go to
TABLE 2: TOP 10 SOCIAL NETWORKING SITES IN THE WORLD email account and click on, forget password’. And the
system will ask a security question like–‘your favorite
Rank Social Networking Estimated Unique Monthly
Sites Visitors teacher, your first phone number etc’. Unknowingly,
1 Facebook 1550,000,000 you have already shared this information with them. So
2 Twitter 94,800,000 now they have access to your email account. This task
3 MySpace 82,500,000 will be much easier if you set your date of birth, school
4 Linkedin 51,000,000 or college name as email password. A study conducted
5 Ning 46,000,000
6 Tagged 38,000,000
by Internet security firm Webroot, found 2 in 10 people
7 Classmates 22,000,000 use significant date, such as birth date, or a pet’s name
8 Hi5 26,000,000 as password which is often publicly visible on social
9 myyearbook 10,000,000 networks. And 4 in 10 respondents shared password
10 Meetup 9,000,000
with at least one person (Boulder, 2010).
Similarly, the Research done by BitDefender
II. IMPACT OF SOCIAL NETWORKING SITES reveals that 75 percent of individuals use same
There is a narrow gap between private and public password for social networking sites and Email (‘Study
life in the age of social networking. Once you put reveals 75,’ 2010). All this ignorance by users becomes
personal data, image or video on web site, you are the main reason behind hacking of SNS profile and
helpless to control its distribution. Though you can set email account.
its privacy level in the profile but you are sharing them B. Effect on Biological Factors
with an unknown web administrator. Tomorrow your
personal information may lend in the hand of parents, Social networking users face severe health risk
teacher, employer, spouse, criminal or a marketing because they reduce face-to-face contact and become
company. A study conducted by Pew Internet & addicted in a virtual world of relationships. Instant
American Life Project in 2007 found that 66 percent of gratification of needs become their goal. According to
teens restrict their online profile by making it private. the U.S. Internet activity (January 2010) Nielsen Online
says, users spent an average of 9 hours a month on
They found 82 percent teens post their first name,
Facebook. This makes Facebook Internet’s maximum
followed by photos of themselves (79%), name of city
time waster. In U.S. 65 percent of people check their
(61%) and email address (29%). According to Pew
Facebook profile before getting out of bed in the
study girls generally don’t post information that could morning and 25 percent check their accounts several
help in finding their physical location. On the other times in a day.
hand, boys are just opposite and post more information
about their real location (Lenhart, & Madden, 2007). C. How Useful to Hackers/ Cybercriminal
A research by Gross and Acquisti after analyzing
The beginning of social networking sites started to
4000 Facebook profiles of Carnegie Mellon University make contacts with people and build a network of
students, found threats emerging from personal healthy relationship in the society. But now it seems
information posted in social networking site. Only 1.2 offering cybercriminals a great advantage to target
percent of users changed their default privacy victims. Haydn and his family were out of home for
preferences. Hence fully identifiable information like treatment of his son in Leeds. Instead of making phone
(first name, last name, personal image, hometown, date calls to everyone, he was updating on Facebook
of birth, ZIP code) easily available to anybody message about son’s health. Someone read his
registered at Facebook. This information can be used to Facebook message and burgled their home during this
estimate a person’s social security number and finding time. A report by Daily Mail shows ‘crimes associated
sensitive medical information (Gross, & Acquisti, with the networking site have increased by as much as
2005). Using the information available on Facebook 7000 percent in some area–including cases of murder,
rape, peddophilia, bulllying, assault and burgglary’ Anallysis of the daata tells us thaat 98% of the students
s use
(‘Facebookk–crime,’ 20110). Pedophilees are using social
s at leaast one of the networking sites.
networkingg sites because that is wherre they can lurre the
0% 0%
victim willlingly. Teens easily believee in Love witth the
person theyy met online but
b actually thhey are fallingg to a
trap of sexuual predator. 4
41% Male
59% Female
Fig. 2
Fig. 1 14
Janis Wolak of the Crimes against Chiildren 12
likedIn
Research Center
C at the University ofo New Hamppshire 10
says: ‘majajority of cases involve young teenaagers, 8
mostly 13--, 14-, 15-yeaar-old girls whho are targeteed by 6 twitter
adults on Internet
I who are straightfoorward about being
b 4
interested in sex’ (Steenhuysen, 20008). In a siimilar 2
facebook
case, Ashleeigh Hall, a 17
1 year old giirl, accepted friend
f 0
request froom a 33 year old man on Facebook.
F Shee had 0 2010 2012
1990 2000
h was a connvicted rapist who
no idea thhat actually he
created a Fake profilee as a teenager. And Finnally, Fig. 3
Ashleigh wasw murdered by that peerson (‘Facebbook–
crime,’ 20010). The majjor problem is i that there is no The number of hours speent on these networking
way to verrify the actuall identity on cyberspace
c annd the sites should be annalyzed here too determine what
w impact
criminals takes the advanntage of this. it woould have onn the users. T The more the time spent,
moree probability ofo these sites being beneficcial is there.
D. Factorrs that Providde Sense that how
h to Safe he majority
m of thhe students sspend on an average of
abouut 6 to 9 hourrs every day ssurfing these networking
When we are usingg these sociaal networking sites sites. Amongst thoose about 28% % users spend more than1
daily theree may be chaance of attackk on us thesse are hourr everyday, raanging betweeen 10 to oveer 15 hours
some of thhe points thatt makes you and your faamily beingg spent every week. The neext question th hat arises is
safe on thhe networkingg sites. whatt else these stuudents would be doing if neetsurfing on
• Avoid acceptinng friend reqquest to unknnown
A thesee sites durinng that time.. When we asked that
peerson. quesstion during thhe survey, wee arrived at a number of
• A
Avoid putting your
y exact infformation ontto the suchh things.
Soocial networking sites.
• Taake precautionns while uploaading photo. A. The Impact onn Youth
• U
Update your prrofile privacy. The impact on youth cann also be calculated by
• D not post neggative things about
Do a your life
fe. analyyzing if these people meet personally to anyone
throuugh these netwworking sitess about 29% ofo the users
III. ANALYSIS
N AND INTERPRETAT
TION SYSTEM havee met someonee personally tthrough these networking
sites which is a huuge number aas it takes a lo
ot of trust in
The popularity
p of the social networking sittes is
meetting someone personally. H Here, the queestions arise
immense, especially ammongst the youth.
y The saample
regarrding safety and
a trust as a lot of peoplee fake their
chosen hadd a majority of students from the Schhools.
profi
files. For som
me, these netwworking sites make them
Bhagwant Institute
e of Technologyy, Muzaffarnaga h, India 77
ar, Uttar Pradesh
Proceedings of National Conference on Trends in Signal Processing & Communication (TSPC’13) 12th–14th April 2013
feel comfortable in their sexuality. For 49%, it is easier 20% of the sample population feels that there is
for them to be themselves while interacting through negative impact on their personal lives.
these sites. It raises question as to how real they are Benefits Users
while they act with people impersonal or in their normal More friends 25%
routine life. Better connectivity 63%
Found my life partner 1%
Enjoy leisure or time when I am alone 22%
Link with prospective recruiters 7%
0%
Who met
28%
throughnetworking sites
V. CONCLUSION
Who met Do you still think social networking sites are, at
throughnetworking sites
72% worst nothing to worry about, at best wholly positive
influences on humanity? If so then I’ll be amazed, but I
would love to hear your reasoning about these negative
Fig. 4: Analysis of Persons Meet on Networking Sites impacts of social networking sites. Will any of this
make me delete my Facebook and Twitter accounts? Of
B. Benefits of Social Networking Sites course not, but it does make me a little more cautious
Better connectivity is what these sites definitely about how (and how often) I use them.
provide. Also people seem to have more friends through
their use. It happens due to finding out old friends and REFERENCES
also remaining in close connection with the old as well [1] http://www.scribd.com/doc/13653301/The-Effect-of-Social-
as the existing friends. This also gives its users a way to Networking-Sites.
enjoy themselves or the leisure time when they are [2] http://www.ebizmba.com/articles/social-networking-websites
alone. They can be themselves and also remain in touch [3] http://www.srikumar.com/homepage/sangeetha_sridhar/social_n
etworking_cyber_crimes-part1.htm
with friends at the same time. Another benefit is the [4] http://article.sapub.org/10.5923.j.ijit.20120101.02.html
opportunity to connect with the prospective recruiters [5] Wright, Alex. (2007, December 2). Friending, ancient or
and to find employment through these sites. Only an otherwise. Retrieved from.
exception of 1% found their life partners through this. www.nytimes.com/2007/12/02/weekinreview/02wright.html
As per the survey, 80% of the people find that there is
no as such negative impacting their personal life. Only
Abstract—This paper describes the use of high K upon the dielectric constant of the medium in which
dielectric materials for better subthreshold performance
it is getting leaked. The fringing field increases with
exhibited by Multi gate MOSFET structures. The
advancement in fabrication technology has also boosted the use of high k dielectric material. The high K
the use of different high K dielectric materials as oxide dielectric material is utilized at different places in
layer at different places in MOSFET structures. One of different MOS devices for performance
the most important multigate structure is FinFET which
improvement[9]. The use of spacer increases the
shown better subthreshold swing (SS) and drain-induced
barrier lowing (DIBL), which is possible with optimum fringing electric field in the FinFET. Apart from
ratio of gate length to Fin width. The SCEs parameters scaling the speed, power requirement of the circuit
such as on-state current (ION), off-state current (IOFF), can be enhanced by using some technique such as
subthreshold slope (SS) and drain induced barrier
using spacers in the FinFETs. Spacers are the
lowering (DIBL) has been drawn. Sentaurus TCAD has
been used for device simulation. dielectric materials used alongside of the gate. Use of
spacer in a FinFET results in increased fringing
Keywords: SOI FinFET, Short channel effect, DIBL,
Subthreshold Slope, High K dielectric material, Spacer. electric field which originates from the sidewalls of
the spacer to the fin region. Since the fringing field
I. INTRODUCTION depends upon the dielectric of the medium in which
the field is getting leaked, so the dielectric material
Multi-gate MOSFET is a promising device for
used for the spacer is important in deciding the
ultimate CMOS device structure because the device
magnitude of the fringing field. This is studied by
shows reduced short channel effect (SCE), higher
using different materials such as Silicon Oxide (SiO2),
current drivability, nearly ideal subthreshold swing
Silicon Nitride (Si3N4) and Hafnium Oxide (HfO2)
(SS), and mobility enhancement[1],[2]. Suppression
having dielectric constants of 3.9, 7.5 and 25
of short-channel effects (SCEs) continues to be a major
respectively. The adoption of relatively thinner spacers
challenge when devices are scaled down to these
of higher permittivity material is concluded to be
dimensions. It was found that as the distance between
beneficial to device performance due to the
source and drain decreases, the gate loses its ability to
suppression of SCE[8].
control the potential distribution and current flow in
the channel, and other undesirable effects, called the
II. DEVICE STRUCTURE AND DIMENSIONS
short channel effects [3] start plaguing MOSFET’s.
This was so because the electric field lies from A FinFET with 20nm channel length is used for
source and drain were affecting the channel. In multi- the analysis. The channel of the FinFET is lightly
gate devices, the FINFET[4] is most suitable structure doped with the concentration of 1e15. The doping level
because it can have simple self-aligned DG and TG in source, drain region is 1e20; Gaussian doping is
used in extension regions. The gate electrode thickness
structure with good process compatibility and easier
used is 40nm while the spacer width is kept at 15 nm.
thickness control of FIN body. Significant scattering of
The fin thickness used for the FinFET is 10nm. The
carriers [5]–[7], increased parasitic resistance due to
Spacer materials used are silicon oxide (SiO2), silicon
thin and narrow source/drain (S/D) regions, and
nitride (Si3N4) and hafnium oxide (HfO2) having
increase in capacitance due to fringing fields in very dielectric constant of 3.9, 7.5 and 25 respectively. The
small 3-D geometry all have an adverse impact on the threshold voltage of the device is 0.2 V. All the
FinFET performance. To fully capture the effect of simulations used in the paper are performed using the
fringing field on FinFET device performance, one has Sentaurus TCAD tools [10].
to consider the non planar nature of the device
geometry and account for the 3-D fringing fields[8].
The magnitude of the fringing electric field depends
ISBN: 978-93-82880-20-2 79
Proceedings of National Conference on Trends in Signal Processing & Communication (TSPC’13) 12th–14th April 2013
IV. CONCLUSION
The use of high k dielectric materials such as
Si3N4, HfO2 improves the strength of the fringing
electric field resulting in increase in the on state
current while the off state current, sub threshold slope
and drain induced barrier lowering of the FinFET is
Fig. 2: Transfer Characteristics of the FinFET with Different decreased, thus resulting in improvement of the device.
Dielectric Materials at VDS=Vdd FinFET with high K dielectric material increases
The variation of on-state current (ION) and off Ion/Ioff ratio which increases the scope for high Wfin
state leakage current (IOFF) with different dielectric (Fig.10). Further, high K dielectric material can be
materials is shown in Fig.3. Fig.4 shows that the useful in double metal gate MOS structures. The
Ion/Ioff ratio increases with increasing dielectric proposed device has improved subthreshold
constant. . The value of sub threshold slope should be performance and avoid the use of underlap behavior of
as minimum as possible so as to have a faster switching gate or channel doping high.
operation and also drain induced barrier lowering
should be as minimum as possible for the better REFERENCES
performance of the device. The sub threshold slope as [1] Y.K. Choi, et al., IEDM 2001, pp. 421–424.
well as DIBL is minimum for Hafnium Oxide spacer [2] J. Kedzierski, et al., IEDM 2001, pp. 437–440.
FinFET while it is maximum for FinFET without [3] J.P. Colinge, FinFETs and other Multi-Gate Transistors
Springer, 2007.
spacer. very essential for low power operation of the [4] Jean-Pierre Colinge, “Multiple-gate SOI MOSFETs” Solid-
device. The decrease in off state current is basically State Electronics 48 (2004) 897–905.
because of increase in the barrier potential faced by the [5] F. Gamiz and M.V. Fischetti, “Monte Carlo simulation of
carriers as explained earlier. Here, the on state current doublegate silicon-on-insulator inversion layers: The role of
volume inversion,” J. Appl. Phys., vol. 89, no. 10, pp. 5478–
has doubled for Hafnium Oxide spacer FinFET if it is 5487, 2001.
[6] A. Khakifirooz and D. A. Antoniadis, “On the electron mobility [9] Saji Josepha, George James Ta and Vincent Mathew “Transport
in ultrathin SOI and GOI,” IEEE Electron Device Lett., vol. 25, characteristics and subthreshold behavior of High-Κ dielectric
no. 2, pp. 80–82, Feb. 2004. double gate mosfets with parallel connected gates” Journal of
[7] M. M. Chowdhury and J. G. Fossum, “Physical insights on Electron Devices, 16, 1363-1369(2012).
electron mobility in contemporary FinFETs,” IEEE Electron [10] “Sentaurus Structure Editor User’s Manual”, Synopsys
Device Lett., vol. 27, no. 6, pp. 482–485, Jun. 2006. International.
[8] M.J. Kumar, S. K. Gupta and V. Venkatraman “compact [11] International Technology Roadmap for Semiconductors (ITRS),
modeling of the effects of parasitic internal fringe capacitance 2007 Edition.
on the threshold voltage of high k dielectric nanoscale SOI
MOSFETs. IEEE transaction on Electronic Devices, 52, 706-
711(2004).
Abstract—The goal of this paper is to discuss about barrier being removed and allowing electrons to tunnel
the basic physics and applications of nano electronic through the junctions and between the source and
device ‘Single electron transistor [SET]’ which is capable the drain.
of controlling the transport of only one electron. Single-
electron transistor (SET) is a key element of current
research area of nanotechnology which can offer low
power consumption and high operating speed. The single
electron transistor is a new type of switching device that
uses controlled electron tunneling to amplify current.
Recent research in SET gives new ideas which are going to
revolutionize the random access memory and digital data
storage technologies.
Keywords: Nanoelectronics, Single-electron transistor;
Coulomb blockade, Coulomb oscillation, Quantum dot
82 ISBN: 978-93-82880-20-2
A Review Paper on Single Electron Transistor and its Applications
one excess electron have equal energies, removing the coupling between qubits but increases the signal-to-
coulomb barrier and allowing tunneling to occur. This noise ratio for both dc and radio frequency single-shot
maximum point occurs when the gate is charged with measurements. Since these results are independent of
exactly minus half an electron. When another minus the device materials, they are useful for establishing
half an electron charge is put on the gate, the coulomb guidelines for the design of SET charge sensors in
barrier is reinstated, resulting in an oscillation in lateral QD-SET structures based on a two-dimensional
conductance of the transistor with maxima at half electron gas.
integer multiples of e and minima at integer multiples
of e. This conductance oscillation allows the single C. Ultrasensitive Microwave Dectector
electron transistor to be used either as a transistor or as Another application of Single Electron Transistor
an extremely precise device for measuring charge. can be as an Ultrasensitive Microwave Detector; island
There are a variety of materials chosen for single is weakly coupled to a bias circuit through two small-
electron transistors based on the particular properties capacitance tunnel junctions and a capacitive gate. At
desired in the system. Relevant properties include the low bias voltages and temperatures, a single
capacitance of the material, the ease of fabrication, quasiparticle may only be introduced to the island
crystalline structure, electron mobility, and ease of through photon-assisted tunneling. Once this occurs, the
growing oxide layers. There are two classes of single quasiparticle is trapped on the island because it takes a
electron transistors used today, ‘metallic " and relatively long time for this specific quasiparticle to
‘semiconducting’. This refers to the material they are tunnel off. While it is trapped, charge is transported
commonly fabricated from as opposed to describing in through the system two electrons at a time. Since the
any way their operation. Both function through the photon-assisted transition merely switches the detector
process of tunneling junctions. current on, this device is not limited to one electron
tunneled through the system per absorbed photon. This
II. APPLICATIONS OF SET makes the device an extremely sensitive and potentially
A. Detection of Infrared Radiation useful detector of microwave radiation.
The high sensitivity of single-electron transistors transistor and measure the gate voltages providing the
have enabled to use them as electrometers in unique sharp increase of the source-drain conductance.
physical experiments. For example, they have made
possible unambiguous observations of the parity effects III. CONCLUSION
in superconductors. Absolute measurements of
This research paper focuses the theoretical
extremely low dc currents (~10-20 A) have been
demonstrated. The transistors have also been used in the discussion of basic principle of Single electron
first measurements of single-electron effects in single- transistor and its applications and importance in the age
of nanotechnology to provide low power consumption
electron boxes and traps. A modified version of the
transistor has been used for the first proof of the and high operating speed in the field of ULSI design for
existence of fractional-charge excitations in the the fabrication of various electronic devices.
fractional quantum hall effect.
REFERENCES
F. Single-Electron Spectroscopy [1] Quantum information technology based on single electron
dynamics, NIT basic research laboratories Atsugi-shi, 243-0198
Another application of single-electron electrometry Japan, Vol. 1 No.3 June 2003.
is the possibility of measuring the electron addition [2] A. E. Hanna and M. Tinkham (1991). ‘Variation of the Coulomb
energies (and hence the energy level distribution) in staircase in a two-junction system by fractional electron charge.’
quantum dots and other nanoscale objects. Physical Review B 44: 5919.
[3] Amiza Rasmi & Uda Hashim ‘Single-electron transistor (SET):
There are two natural ways to carry out such Literature Review’ journal 2005, koieg University, Malaysia.
measurements. The first is to use the quantum dot as the [4] Andreas Scholze, ‘Simulation of single-electron devices,’ Ph.D.
island of the single-electron box, capacitively coupled dissertation, Univ. of Jena, Germany, 2000.
to the single electron transistor or other sensitive [5] Gregory S. Dubejsky, ‘Fabrication and DC characterization of
Single electron transfers at low temperature’, M.S. Thesis,
electrometer. The second is to use the quantum dot Queen’s Univ. Kingston, Ontario, Canada, Aug. 2007.
directly as the island of a weakly biased single-electron
ISBN: 978-9
93-82880-20-2 85
Proceedingss of National Co
onference on Tre
ends in Signal Processing
P & Co
ommunication (TSPC’13), 12th
h–14th April 2013
II. PRELIMINARIES
R
86 Bhagw
want Institute of Technology, Muzaffarnagar, Uttar
U Pradesh, In
ndia
Protection of Embedded Software from Side Channel Attacks using Virtual Secure Circuit
X. EXPERIMENTAL SETUP
The experimental setup for power attacks used
an automatic measurement system that contained a
FPGA board, an oscilloscope and a PC, shown in
Figure 5. A RS232 cable connected the FPGA board
and the PC to enable the communication between these
two. To do power attacks, the oscilloscope used a
current probe (Tektronix CT-2) to monitor the current
flowing into the FPGA core. Here, current represents
the power of the FPGA. In the electromagnetic attacks,
Fig. 4: Sample Mentor Graphics ModelSim Waveform for the the oscilloscope used a H-field probe (ETS-LINDREN
Proposed NCL S-Box with Input Signals Changing from 0d to 3d 903) to detect the magnetic field. Between the
TABLE I. SIMULATION RESULTS FOR TEN ARBITRARY SAMPLES FROM oscilloscope and the PC, communication was done
THE CONVENTIONAL SYNCHRONOUS S-BOX AND THE PROPOSED NCL through a USB cable. We programmed the PC to send
S-BOX command to the oscilloscope and plaintext to the
FPGA, and to obtain the sampled signal from the
oscilloscope, after one sampling is done.To reduce
noise sampled by probes, every sampled trace
transferred to the PC is an average trace of 32
normaltraces (with the same plaintext).
use of Hamming weight of only 1 bit of the S-Box’s SCA resistance. Also, DRP technique is considered to
output. Usually, the byte-based model is more efficient have a good nature against fault attacks. Integrating
because it contains more useful information. However, fault-attack resistance to VSC also deserves further
when attacking DRP countermeasures, sometimes the investigation. As for future deployment, we see the
bit-based model could be better. This is because the first step to deploy VSC as a practical SCA solution on
imbalance within complementary wire pairs could be extensible processors.
positive (e.g. the direct wire is longer) or negative (e.g.
the complementary wire is longer), depending on REFERENCES
specific circuit routing. In such case, adding Hamming [1] M.-L. Akkar and C. Giraud, “An Implementation of DES and
weights of different bits in a byte together may average AES, Secure against Some Attacks,” CHES 2001, vol. LNCS
out the imbalance, which is not helpful for attacks. The 2162, pp. 309–318, 2001.
experiments tried both models and selected the more [2] P.C. Kocher, J. Jaffe, and B. Jun, “Differential Power
Analysis,” CRYPTO 1999, vol. LNCS 1666, pp. 388–397,
powerful one to present the results. 1999.
[3] M. Lazzaroni, V. Piuri, and C. Maziero, “Computer security
XII. CONCLUSION AND FUTURE WORK aspects in industrial instrumentation and measurements,” in
Proc. IEEE I2MTC, May 2010, pp. 1216–1221.
This work proposes VSC as a solution to protect [4] J. Kocher, P. Jaffe, and B. Jun, “Introduction to differential
software from SCA with the DRP technique. This power analysis and related attacks,” Cryptography Res. Inc.,
San Francisco, CA, 1998, Tech. Rep.
concept can be implemented following two schemes. [5] D. Macii and D. Petri, “Accurate software-related average
One is on an customizable single-core processor and current drain measurements in embedded systems,” IEEE
the other is on a synchronized dual-core processor. A Trans. Instrum. Meas., vol. 56, no. 3, pp. 723–730, Jun. 2007.
set of real-world attacks based on power and [6] V. Konstantakos, K. Kosmatopoulos, S. Nikolaidis, and T.
Laopoulos, “Measurement of power consumption in digital
electromagnetic emanation have been done to quantify systems,” IEEE Trans. Instrum. Meas., vol. 55, no. 5, pp.
the security improvements. Except that dual-core VSC 1662–1670, Oct. 2006.
is still breakable by electromagnetic attacks, in all the [7] D. Macii and D. Petri, “An effective power consumption
other cases, the VSC schemes offer comparable measurement procedure for bluetooth wireless modules,” IEEE
Trans. Instrum. Meas.,vol. 56, no. 4, pp. 1355–1364, Aug.
protection against power and electromagnetic attacks 2007.
when compared with the real DRP circuits. In addition, [8] L. Angrisani, M. D’Apuzzo, and M. Vadursi, “Power
considering its algorithm independent and low- measurement in digital wireless communication systems
hardware-cost features, we expect VSC an effective through parametric spectral estimation,” IEEE Trans. Instrum.
Meas., vol. 55, no. 4, pp. 1051–1058, Aug. 2006.
SCA countermeasure for software cryptography. [9] T. Lopez and R. Elferich, “Measurement technique for the static
Symmetric routing is important to DRP output characterization of high-current power MOSFETS,”
technique. A further step to this work can be IEEE Trans. Instrum. Meas., vol. 56, no. 4, pp. 1347–1354,
combining VSC and symmetric routing for higher Aug. 2007.
Abstract—Process support systems, such as resources and the control of the flow of the associated
workflows, are being used in a variety of domains. information required by performers to execute the
However, most areas of application have focused on tasks. Typically the tasks involved in the business
traditional production-style processes, which are process are interdependent in that the execution of
characterized by predictability and repetitiveness.
Application in non-traditional domains with highly
one task is conditional upon the execution of one or
flexible process is still largely unexplored. Such flexible a number of other tasks. Workflow management
processes are characterized by lack of ability to completely systems employ a process model to capture this flow
predefine or an explosive number of alternatives. In this of execution between tasks. This model is used by a
paper, we will present an approach to building workflow workflow management system to schedule and
models for such processes. We will present our approach coordinate the execution of these tasks. Production
in the context of a non-traditional domain for workflow workflows based on this framework have been
deployment, which is, degree programs in tertiary developed to automate the coordination of the activities
institutes. The primary motivation behind our approach is for processes that are typically characterized by well-
to provide the ability to model flexible processes without
introducing non-standard modeling constructs. This
defined procedures and are highly repetitive in nature.
ensures that the correctness and verification of the The processes may be complicated in nature,
language is preserved. We propose to build workflow involving a large number of tasks, performers and
schemas from a standard set of modeling constructs and coordination constraints. However many new
given process constraints. We identify the fundamental categories of workflow technology are emerging to
requirements for constraint specification and classify them address the diverse range of processes now looking to
into selection, termination and build constraints. Finally, engage the technology due to the automated
we will demonstrate the dynamic building of instance coordination benefits that it provides.
specific workflow models on the basis of these constraints.
The ability to predefine a business process
Keywords: flexible workflow, tertiary education,
constraint specification
completely cannot be relied upon. Furthermore even
highly predictable processes are subject to change as
I. INTRODUCTION organizations adjust their activities in response to
influences such as new legislation, innovations and
Effective business organizations develop processes competitive pressures. Another pressure on the rigid
to ensure that their activities are performed in a process definition comes from the need for support of
consistent and reliable manner. These processes exceptional cases, in particular unforeseen scenarios
consist of four primary components, namely objects, that cannot be suitably addressed by the existing
tasks, performers and constraints. The object is a process definition. At the other extreme of a continuum
business entity, such as a home loan application, of support for process change is the complete
which is required to be worked on during its life relaxation of coordination constraints, which has lead
cycle. The business activities defining this work form to developments in ad-hoc workflows to cater for
the tasks to be performed, such as checking the complete flexibility in the execution. Positioned
applicant’s credit rating or current financial between the extremes of this continuum is the support
commitments. Performers carry out these tasks on the for processes that can be partially defined but not
objects, as the task cannot perform its associated completely specified until runtime.
work on the object directly. A performer may be a A typical example is healthcare, where patient
person, typically someone filling an organizational admission procedures are predictable and repetitive,
position, such as a manager or a supervisor, or it however, in-patient treatments are prescribed uniquely
may be for each case, but none-the-less have to be coordinated
Workflow systems are currently the leading and controlled.
technology for supporting business processes.This Another application is higher education, where
technology manages the execution of the tasks students with diverse learning needs and styles are
involved in a business activity, the scheduling of working towards a common goal (degree). Study paths
90 ISBN: 978-93-82880-20-2
On Capturing Exceptions in Workflow Models for Flexible Processes
taken by each student need to remain flexible to a large we will utilize only the basic set of modeling
extent, at the same time providing study guidelines and constructs, consisting of Sequence, Exclusive Or Split
enforcing course level constraints is necessary to ensure (Choice) Exclusive Or Join (Merge), And Split (Fork),
a certain quality of learning. And Join (Synchronizer). The graphical representation
Web content management is also characterized by of these constructs is shown in Figure 1.
flexible processes, where especially in large projects,
every development suggests the need for an overall
plan to provide the objectives, approvals, and strategy,
as well as a flexible means of coordinating the
combined efforts of the theme designers, graphic
experts, programmers, and project planners.
Effective Customer Relationship Management
(CRM), a critical component in enterprise solutions,
also signifies the need to provide a flexible means of
composing call center activities according to the
available resources and data, but within certain Fig. 1: Basic Workflow Constructs
constraints, thus allowing for an integration of CRM The semantic context of the basic constructs is
systems with core organizational workflow processes presented with reference to the process depicted in
and underlying applications. Figure 1.
We will introduce in the following section, some Sequence enables an activity in the workflow after
basic terminology and a simple workflow modeling the completion of another activity in the same process
language. In the next section we will present an as shown by activity C’s dependency on Activity B.
analysis of flexible processes in the context of And Split (Fork) supports simultaneous execution
tertiary degree processes, and identify the fundamental in the process by allowing a single thread of control
requirements for constraint specification for such to split into multiple threads that can be executed in
processes. We will also demonstrate the unsuitability of parallel. The branch containing Activities B, C and D
‘prescriptive’ process modeling languages in this area. can execute simultaneously with the branch containing
Finally, we will introduce our proposed approach. We Activities E and F.
will show how flexibility of specification can be And Join (Synchronizer) brings together multiple
achieved without compromising the simplicity and parallel threads into a single thread of control. Activity
consequently verifiability of the modeling language. G will only trigger after the incoming transitions from
the branch containing Activity B, C and D and the
II. BASIC TERMINOLOGY branch containing Activity E and F are triggered.
We will use basic workflow modeling concepts to Exclusive Or Split (Choice) represents a point in
demonstrate our approach. Below we introduce basic the workflow process where one of several branches is
terminology for the sake of clarity: chosen based upon the results of a condition. That is,
• We define a workflow schema as a Activity E or Activity F will follow Activity A in this
representation of a set of activities, the branch of the process.
ordering and interdependencies between these Exclusive Or Join (Merge) allow alternate
activities, the resources available to perform branches to come together without synchronization.
them and their information or data Completion of either Activity E or Activity F will
requirements that are required to complete a trigger the merge.
business process. The above defines the structural layer of the
• A workflow instance, like its database workflow that captures the flow of execution from one
namesake, denotes a particular occurrence of task to another. In addition a temporal layer is required
the business process as defined by the to capture scheduling constraints additional to those
schema. imposed by the structure, a data layer to capture the
• An instance type is the set of instances that data requirements of workflow tasks, and data flow
follow the same execution path through the between tasks, and an execution layer to schedule and
conditions that exist in the workflow schema. control the actual tasks invocations, resources and data.
Therefore, each task in the workflow is described by a
A. Semantics of the Workflow Language set of properties. These properties relate to the data,
We also introduce a graphical workflow definition time, underlying applications, resources, clients,
language, which will be used for demonstrating various compensation and much more. We do not elaborate on
concepts. This language conforms closely to the task properties in this paper. However, the task is a
workflow management coalition standards. Note that complex object with rich semantics, and cannot be
considered as a mere node in the workflow graph.
chosen domain. However, at a more generic level, we These dependencies can be depicted as a control
see the use of structures that allow multiple executions flow structure in the given workflow language, as
in sequence and/ or in parallel. While choice and merge shown in Figure 3.
constructs may be present within workflow fragments,
we propose that these constructs not be used to build the B. Termination Constraints
instances. Since an instance represents a particular In the absence of an explicit termination task, it is
occurrence of the workflow process, the choices should essential to identify the termination constraints, which
be made during the building of the instance. The define to a large extent the process goal. Furthermore,
elimination of the choice-merge construct from the these constraints must be defined in terms of the
instance template further has the advantage of fragments from which the instances are built. In the
simplifying the model, and removing the chance of higher education domain, we can identify the following
deadlocks or lack of synchronization. termination constraint;
A set of constraints that will define the rules The total number of courses (units) to be
under which valid instances can be built. We identify completed is always specified but some programs may
three levels of constraint specification. require that a specified number of units be completed
• Selection Constraints. in a particular area or at a particular level or a
• Termination Constraints. combination of both.
• Build Constraints.
C. Build Constraints
A. Selection Constraints
In addition to the rules captured by the
Selection constraints dictate the inclusion of a selection and the termination constraints, further
fragment in the pool of fragments available for restrictions exist to control the building of the
building. In degree programs, fragments are simply process. The build constraints specific to this domain
courses. A course is available for selection (included are typically:
in the pool of fragments available for building) when • Rules specifying the availability of course
certain constraints have been met. In this domain we (semester 2, alternate years).
identified three types of constraints which are basically • Rules specifying the minimum and maximum
interdependencies that exist between courses. These student workloads (units taken per semester).
are prerequisite, companion and incompatible. • Rules specifying performance requirements.
Prerequisite requirements for a course describe its For example, entry into honors programs may
dependency on the completion of a set of courses. require an overall grade average or a grade
These prerequisite requirements may take the form of average taken over a number of courses.
satisfactory completion of:
• A specific course. V. BUILDING VALID INSTANCES
• A number of specific courses.
• A number of subjects in an area of study or The dynamic build approach allows for the
at a study level or in an area of study at continuous build of an instance from the set of
a level. available fragments and modeling constructs. The
absence of a constraint violation permits the build to
• Advanced standing
continue. Initially the instance specification will be
One of a number of alternatives, where each
undefined and the pool of fragments available to the
alternative may be in any one of the above forms.
build will contain only those fragments without
Companion requirements describe courses that are
selection constraints. Specification of the instance can
dependent upon other courses being undertaken in
begin with a selection of the available fragments
concurrence with them.
connected through the constructs to form what we call
the open instance. The instance will remain open
until such times as the fragments it is comprised of
satisfy the termination constraints of the process. The
closed instance specification therefore defines a valid
instance type that may be adopted by future instances
of the process. Building of the open instance is
progressive and dynamic. The set of fragments
available to the build activity, at a point of time in the
Fig. 3: Higher Education Domain Task Interdependencies build, is dependent upon the fragments that exist in
the open instance at that point of time. As fragments
Finally a course may be incompatible with another are added to the open instance more fragments have their
course. Successful completion of an incompatible for a selection constraints satisfied by it and are added to the
course will prevent the credit for that course.
pool of available fragments. This is demonstrated in checks on the instance data of the open instance under
Figure 4 by an example of a set of five fragments of development. Once the instance data check identifies
which two are available through the satisfaction of their that a termination constraint has been satisfied, the
selection constraints by the fragments contained in the instance is allowed to complete. A simple example is
open instance. In the figure it should be noted that the presented in Figure 6. In the example the termination
selection constraints for each fragment have themselves constraint is satisfied if any four fragments from the
been represented in the workflow modeling language for set Fragments (A, B, C, D, E) are successfully
clarity of the example. completed.
requirements that recur frequently in business paper provides a comprehensive analysis of flexible
processes. Given the fundamental differences this workflow processes derived from an analysis of the
introduces to a workflow management system, use of higher education domain. It attempts to address this
these language extensions will have implementation flexibility by developing a framework for specifying the
consequences resulting in the loss of genericity and process model that can be tailored to individual
limited verification support. instance requirements. Our basic idea is to provide a
Within well-defined fixed processes the need to means of capturing the logic of highly flexible
support dynamic change as the business process processes without compromising the simplicity and
evolves in response to competitive and regulatory genericity of the workflow specification language or
influence, still exists. The biggest problem is the seriously impact the functionality of the underlying
handling of the active instances that were initiated in workflow engine. The framework proposed will provide
the old model. Defining a migration strategy is a an approach to modeling dynamic organizational
complex problem and has been the target of extensive processes that avoids the overheads incurred in their
research. Furthermore the need to handle exceptions to specification, and modification in response to change,
cater for instances that cannot be anticipated at design that arise when a prescriptive approach is taken.
must also be supported. As the support for this level of
REFERENCES
change is required in the traditional workflow
implementations considerable work has been devoted [1] Abraham Bernstein, How can Cooperative Work Tools Support
to it also. Another aspect of work in this area is based Dynamic Group Processes? Bridging the Specificity Frontier,
CSCW’00, December 2-6, 2000, Philadelphia.
on the concept that an exception may result in a [2] Douglas P. Bogia, simon M. Kaplan. (1995) Flexibility and
change in the level of specificity of the process. For Control for Dynamic Workflows in the wOrlds Environment.
example a highly specified process of processing Proceedings of ACM Conference on Organizational Computing
customer orders may move to a highly unspecified Systems COOCS 95, Milpitas, CA. USA, Nov 1995.
[3] Fabio Casati, S. Ceri, B. Pernici, G. Pozzi. (1996) Workflow
collaborative process involving emails and Evolution. Proceedings of the 15th International Conference
consultations as a result of strike conditions at a on Conceptual Modeling, ER’96, Cottbus, Germany. Springer
particular manufacturing plant. The developments Verlag, Lecture Notes in Computer Science.
supporting these dimensions of change are based on [4] Thomas Herrmann, Evolving Workflows by User-driven
Coordination, Proceedings of DCSCW, Munich, Germany,
the underlying assumption that the change is Sept. 2000, pp. 102–114.
exceptional. In the processes that our work aims to [5] Chengfei Liu, Maria E. Orlowska, Hui Li, Automating
support the change is inherent in the process. Handover in Dynamic Workflow Environments. Proceedings of
Moving to the other end of our continuum for 10th International Conference on Advances in Information
System Engineering (CAiSE 98), Pisa, Italy, June 1998.
organizational processes that spans from highly [6] Markus Kradolfer, Andreas Geppert. (1999) Dynamic Workflow
specified and routine processes to highly unspecified Schema Evolution based on Workflow Type Versioning and
and dynamic processes we acknowledge the Workflow Migration. Proccedings of the Fourth IFCIS
significant work that has been performed in the International Conference on Cooperative Information Systems
(CoopIS99). Edinburgh, Scotland. Sep 2–4, 1999.
coordination of collaboration intensive processes. The [7] Shazia Sadiq (2000) Handling Dynamic Schema Change in
complete relaxation of coordination, to support ad-hoc Process Models. Australian Database Conference, Canberra,
processes is not conducive to the processes targeted by Australia. Jan 27-Feb 02, 2000.
our work. [8] Shazia Sadiq (2000) On Capturing Exceptions in Workflow
Process Models. Proceedings of the 4th International
Positioned between the two extremes of our Conference on Business Information Systems. Poznan, Poland.
continuum are developments in the areas of evolving April 12–13, 2000.
workflows. This work focuses on processes where [9] Wasim Sadiq, Maria Orlowska (1999) On Capturing Process
only part of the procedure can be adequately Requirements of Workflow Based Information Systems. In
Proceedings of the 3rd International Conference on Busi-ness
predefined and the balance is non-procedural content. Information Systems BIS ‘99, Poznan, Poland, April 14–16,
The specification of the selection and sequencing 1999. pp. 195–209. Springer-Verlag, 1999.
of the activities of the unchoreographed content, [10] Sadiq, W. & Orlowska, Applying Graph Reduction Techniques
which involves resources and collaboration, is for Identifying Structural Conflicts in Process Models, 11th
Conf on Advanced Information Systems Engineering,
supplied by the performers at the time the tasks are CAISE'99, Germany, pp. 195–209, 1999.
invoked. This work relates closely to the work [11] Shazia Sadiq, Wasim Sadiq, Maria Orlowska. (2001) Pockets of
presented in this paper but is distinguished from it in Flexibility in Workflow Specifications. (To Appear)
that the unchoreographed content is contained to Proceedings of the 20th International Conference on
specific sections of the predefined process. Conceptual Modeling ER2001. Yokohama, Japan. November
27–30, 2001.
[12] Reichert Manfred, Dadam Peter (1997) Adeptflex -Supporting
VII. CONCLUSION Dynamic Changes of Workflow without loosing control.
[13] Journal of Intelligent Information Systems JIIS, Special Issue
The inability of workflow technology to eloquently on Workflow and Process Management.
model processes containing a high degree of flexibility [14] Joe Lin, Charley Ho, Wasim Sadiq, Maria E. Orlowska (2001)
has limited its deployment in some domains. This On Workflow Enabled e-Learning Services. (To Appear)
ICALT 2001, Madison, Wisconsin, Aug 2001.
Abstract—Wireless power transfer (WPT) via transfer distant were not discussed [7]. However, in
magnetic resonant coupling has been attracting research actual wireless power transmission applications, load
attention for various applications. Conventionally, load is and transmitting distance change frequently.
assumed to be constant and only transfer efficiency is Information such as the ratio of input to output voltage,
studied. In actual WPT applications, the load and transfer the ratio of input to output current and input impedance
distance change frequently. Furthermore, information
such as ratio of input voltage to output voltage, ratio of
seen from high frequency power supply are needed to
input current to output current, and input impedance are understand and construct the power transfer system
[8]
needed for understanding and constructing the power .This paper presents not only the mathematical
transfer system. In this paper, not only the transfer expression for transfer efficiency in SS-type magnetic
efficiency but also the three parameters mentioned are resonant coupling but also the voltage ratio, current
studied. These parameters are then analyzed for changing ratio and input impedance in terms of load and
load and changing transfer distance conditions using transmitting distance change. The derived equations are
actual antenna’s parameters. From the analysis results, investigated using simulations. From these
the load resistance value for maximum efficiency exists. mathematical expressions, important information for
Secondly, improving efficiency by changing load
resistance for small mutual inductance case has larger
wireless power transfer design such as voltage across
effect. The optimum load resistance also changes the load and load impedance values during maximum
according to transmission distance and also the consumed efficiency are obtainable.
power peak may not correspond to maximum efficiency.
Finally, fault protection may also be necessary for the
cases when the load resistance is extremely high and when
the receiver antenna is not present causing high supply
current.
Keywords: Introduction, Self resonance theory,
analysis with variable distance & load and conclusion.
I. INTRODUCTION
Wireless Power Transfer (WPT) via magnetic
resonant coupling which was first introduced in year Fig. 1: Wireless Power Transmission via Resonance
2006 has been receiving much attention from
researchers and companies [1]. With this method high II. CHARACTERISTIC OF SELF
transfer efficiency is obtainable over relatively larger RESONANCE FRQUENCY
gap compared to induction method. Moreover the A. Equivalent circuit
magnetic field in this power transfer method is non-
radiative type and therefore is safe for the human body The resonant antennas used in WPT consist of coils
[2] [3]
. The resonant antennas used in wireless power and capacitors. Antenna design with SS configuration
transfer consist of coils and capacitors. Antenna design where the coils and capacitors are connected in series is
with Series-Series (SS) configuration where the coils preferred in recent research [9]. The equivalent circuit of
and capacitors are connected in series is preferred in magnetic resonance coupling method with SS
recent research[9].Conventional studies assume the load configuration has already been proved in the past
is always constant and discussed the transfer efficiency research [9], and is illustrated in Fig. 2. L1 and L2
only in terms of input power and output power[4]-[6] but represent the inductances of the coils. C1 and C2
not in terms of voltage and input impedance. represent the capacitances that are connected to coils in
Furthermore, the effects of changing load and changing series. R1 and R2 are the antennas’ internal losses. Lm is
96 ISBN: 978-93-82880-20-2
Analysis of Characteristics of Magnetic Resonant Coupling in Wireless Power Transfer
mutual inductance which is related to transfer distance. frequency is usually the same as antenna’s self resonant
Transmitting antenna and receiving antenna satisfy (1). frequency given by (1).
Term ωo in (1) is the self resonant frequency. The V1 Z11 Z12 I1
= (6)
equivalent circuit of the wireless power transfer system 0 Z21 Z22 I2
is shown in Fig. 2. Z11 = R1 + j(ωL1 − ) (7)
ωC
ωo = = (1) Z12 = Z21 = iωLm (8)
√L C √L C
Z22 = R2 + RL + i(ωL2 − ) (9)
ωC
Substituting the self resonant frequency, (2)-(5)
become (10)-(13).Equation (10) and (11) contain only
imaginary component. This shows that voltage and
current are phase shifted by 90 degree from input to
output regardless of the load impedance and
transmitting distance. On the other hand, (12) contains
only real component. This shows that power waveform
Fig. 2: Equivalent of Magnetic Resonance Coupling does not shift phase from input to output. Equation (13)
with Series-Series Configuration
also contains only real component showing that power
B. Definition of the Four Characteristics factor is 100%.At ω=ωo,
ω L RL
Av = j (10)
In this section, the four characteristics used to study R R (ω
R RL ) L
ω L
the phenomena of magnetic resonant coupling are AI = j (11)
RL R
defined. Fig. 3 shows T-type equivalent circuit of the (ω L ) RL
wireless power transfer, and is terminated by a load and Ap = (R (12)
R )(R R R RL (ω L ) )
a power source on each side. Properties to be defined (ω L )
Zin = R1 + (13)
are AV, AI, AP, Zin. Where AV is ratio of output to input RL R
voltage and AI is ratio of output to input current as
shown by (2) and (3). AP is the ratio of output to input III. ANALYSIS WITH VARIABLE DISTANCES AND LOAD
power as shown in (4). AP is the same as transmitting Using the derived equations, the change of each
efficiency. Power ratio is the product of AV and characteristic corresponding to load, RL and
complex conjugate of AI. Zin is the input impedance transmitting distance (which affects mutual inductance
seen from the power source as shown in (5). From (2)- Lm) is calculated. The antennas’ parameters are set to be
(5), the relation of power supply’s voltage and power the same as the actual antennas. L1 and L2 are 800 uH,
supply’s current can be known. C1 and C2 are 2000 pF and R1 and R2 are 1.2 Ω. Table 1
V
Av = (2) shows the Lm values of a few transmitting distances.
V
I
Ai = (3) TABLE 1: LM VS. TRANSFER DISTANCES
I
Ap = Av × Ai (4) S. N. Lm[uH] Distances[cm]
V 1 315 5
Zin = (5)
I 2 240 10
3 172 15
4 102 20
5 76 25
6 49 30
7 38 35
8 26 40
9 20 45
10 15 50
is powered, the voltage at the secondary side increases Fig. 6: Current Gain vs. RL with Different Lm
with the load resistance. This voltage may increase to a
dangerous level without proper control. Fig. 6 shows
that AI decreases with increasing RL. From (4), AP is
the product of AV and AI. Furthermore, from Fig. 4 and
Fig. 6, AP decreases due to decreasing AI when RL is
high whereas AV is saturated at this point. On the other
hand, efficiency declines when RL is extremely low due
to decreasing AV.Fig. 7 show that Zin decreases when
RL increases. The input impedance is low when RL
increases and no load condition occur. Power supply
current, I1 increases. Result of Fig. 5 and Fig. 7 shows
that no load (RL is infinite) may caused damage to both
the load and the power supply.
REFERENCES
[1] Andre Kurs, Aristeidis Karalis, Robert Moffatt, J.D.
Joannopoulos, Peter Fisher, Marin Soljacic,“WPT via Strongly
Fig. 10: AI when Lm Changes Coupled Magnetic Resonances ”,Science, Express, Vol.317,
No.5834, pp.83–86, 7 June 2007.
[2] Hori, Y.;, "Novel EV society based on motor/ capacitor/
wireless—Application of electric motor, supercapacitors, and
wireless power transfer to enhance operation of future vehicles,"
Microwave Workshop Series on Innovative Wireless Power
Transmission: Technologies, Systems, and Applications
(IMWS), 2012 IEEE MTT-S International , vol., no., pp.3-8, 10-
11 May.
[3] Aristeidis Karalis, J.D. Joannopoulos and Marin Soljačić,
“Efficient wireless non-radiative mid-range energy transfer,”
Annals of Physics, Volume 323, Issue 1, January 2008, Pages
34–48, January Special Issue 2008.
[4] Zhen Ning Low; Chinga, R.A.; Tseng, R.; Jenshan Lin; ,
"Design and Test of a High-Power High-Efficiency Loosely
Coupled Planar Wireless Power Transfer System," Industrial
Electronics, IEEE Transactions on , vol.56, no.5, pp.1801–1812,
May 2009.
[5] Sample, A.P.; Meyer, D.A.; Smith, J.R.; , "Analysis,
Fig. 11: Zin when RL Changes
Experimental Results, and Range Adaptation of Magnetically
Fig. 10 shows that input impedance, Zin decreases Coupled Resonators for Wireless Power Transfer," Industrial
Electronics, IEEE Transactions on , vol.58, no.2, pp.544–554,
with lower Lm. For further transmission distance, the Feb. 2011
input impedance is low as if the receiver antenna does [6] Imura, T.; Hori, Y.; , "Maximizing Air Gap and Efficiency of
not exist. Therefore, the power supply current (I1) may Magnetic Resonant Coupling for Wireless Power Transfer
Using Equivalent Circuit and Neumann Formula," Industrial
increase to a dangerous level if there is no fault Electronics,IEEE Transactions on , vol.58, no.10, pp.4746-4752,
protection circuit. Oct. 2011.
[7] Hosotani, T.; Awai, I.; , "A novel analysis of ZVS wireless
power transfer system using coupled resonators," Microwave
IV. CONCLUSION Workshop Series on Innovative Wireless Power Transmission:
Technologies,Systems, and Applications (IMWS), 2012 IEEE
The characteristics of wireless power transfer MTT-S International ,vol., no., pp.235–238, 10–11 May 2012
which are the voltage ratio, current ratio, and input [8] Moriwaki, Y.; Imura, T.; Hori, Y.; , "Basic study on reduction of
impedance and transfer efficiency are studied and reflected power using DC/DC converters in wireless power
calculated using MATLAB coding mathematically. The transfer system via magnetic resonant coupling,"
Telecommunications Energy Conference (INTELEC), 2011
analysis is performed not only for constant load case, IEEE 33rd International , vol., no.,pp.1-5, 9-13 Oct. 2011
but also for changing load and changing transfer distant [9] Imura et al.:"WPT during Displacement Using Electromagnetic
using the parameters of actual antennas. From the Coupling in Resonance -Magnetic versus Electric Type
Antennas-",IEEE Trans. IA, Vol. 130, No. 1, pp.76–83 2010.
analysis results, the optimal load resistance value for
Abstract—In this paper design of stepped Impedance frequency techniques for fabrication does not fit at these
Low Pass Filter for L-band application is presented. frequencies due to the very high losses associated [1-5].
Microwave systems have an enormous impact on modern Although micro strip is not the highest performance
society Micro strip filters play various roles in wireless or filter technology, still it is the preferred choice in many
mobile communication systems. There is an increasing
demand for newer microwave and millimeter-wave
thin-films on ceramic and printed circuit board
systems to meet the emerging telecommunication applications. RF pre-selector filters, image rejection
challenges with respect to size, performance and cost. This filters, local oscillator (LO) filters and intermediate
paper describes the design of low cost and low insertion frequency (IF) filters can all be realized in micro strip.
loss L-band Low Pass Filter (LPF) by using micro strip The recent advance of novel materials and fabrication
line which achieve cut off frequency 2.24 GHz for technologies, including monolithic microwave
permittivity 4.2 value with a substrate thickness 2.4mm integrated circuit (MMIC),micro electro mechanic
for order n=6. Micro strip technology is used for simplicity system (MEMS), micromachining, high-temperature
and ease of fabrication. The design and simulation are Superconductor (HTS), and low-temperature co fired
performed using 3D full wave electromagnetic simulator
IE3D. The stepped-impedance low pass filters have 20dB
ceramics (LTCC), has simulated the rapid development
insertion loss at 3.4 GHz. of new micro strip and other filters [6-7].
In this paper, low pass filter is optimized for high
Keywords: Microstrip filter; L-band; IE3D; Stepped performance and an efficient. Micro strip technology is
impedance; Cutoff Frequency; Return Loss used for simplicity and ease of fabrication. The design
and simulation are performed using 3D full wave
I. INTROUDUCTION electromagnetic simulator IE3D.
Microwave communication systems are expanding
rapidly to higher frequency such as L-band since they II. FILTER DESIGN METHOD
can provide many advantages over conventional The design of low pass filters involves two main
wireless links, for example the larger bandwidth and steps. The first one is to select an appropriate low pass
smaller device size. Waveguide components are widely prototype. The choice of the type of response, including
used at these frequencies and offer very good pass band ripple and the number of reactive elements
performance but the result in high production costs and (order of the filter) will depend on the required
bulky systems. Conventional planar filter structures specifications. The element values of the low pass
suffer from radiation from the resonators into the prototype filters, which are usually normalized to make
substrate and from high ohmic loss, and therefore give a source impedance g0=1 and a cutoff frequency
high insertion loss and poor filter rejection However; Ωc=1.0, are then transformed to the L-C elements for
there is still a challenge in lowering the cost of the the desired cutoff frequency and the desired source
commercial RF micromachined device since the devices impedance, which is normally 50 ohms for micro strip
are fabricated in semiconductor- like process. The filter filters. The next main step in the design of micro strip
is one of the primary and necessary components of a low pass filters [8] is to find an appropriate micro strip
microwave system. Microstrip line is a good candidate realization that approximates the lumped element filter.
for filter design due to its advantages of low cost, The element values for the low pass prototype with
compact size, light weight, planar structure and easy maximally flat response at pass band ripple factor
integration with other components on a single circuit LAR=0.1dB. Characteristic impedance source/load
board. Conventional filter structures like equal ripple Z0=50 ohms, are taken from normalized values gi i.e.g1,
and Butterworth low pass filters are requirement of g2, g3, g4……gn, The filter is assumed to be fabricated
special fabrication methods. Conventional low on a substrate of dielectric constant εr and of thickness
h mm for angular (normalized) cutoff frequency Ωc, TABLE 1: DIMENSIONS OF THE STEPPED IMPEDANCE
LOW PASS FILTER (FOR N=6)
using the element transformation [9].
The filter design steps are as follows: Section Zi= Zl or βl (Degree) Wi(mm) li(mm)
1. Determine the number of sections from the ZH (Ω)
1 20 11.86 11.39 2.1
specification.
2 150 27 0.1789 5.42
Characteristics for micro strip parameters. 3 20 44.29 11.39 7.79
Filter Design Specifications: 4 150 36.9 0.1789 7.4
• Relative Dielectric Constant εr = 4.2. 5 20 34.42 11.39 6.1
6 150 9.89 0.1789 1.98
• Height of substrate, h=2.4mm. 7 50 30 3.04 6.85
• Cutoff frequency=2.5 GHz.
• The loss tangent tanδ=0.02. IV. SIMULATION RESULT
• The filter impedance Z0 =50Ω. For the simulation purpose we have used Full-wave
• The highest line impedance EM solver IE3D.The dimensions of the filter are given
ZH=ZOL=150Ω. in the above table.
• The lowest line impedance Zl=Zoc=20Ω.
Ωc=1.
2. We have taken the element value for low pass
filters from [9] for n=6.
3. The electrical lengths of the inductor
=
The electrical length of the capacito0r sections
Where, A (2)
Where (3)
Where η=120π ohms is the wave impedance in free
space.
5. For
(4)
6. The effective dielecteic constant can be found Fig. 2: Full-Wave EM Simulated Performance of the Stepped
by the following formula [7] Impedance Low Pass Filter n=6 at 2.24GHz
(5)
7. Effictice wavelength is also found as [2]
(6)
V. RESULT & ANALYSIS space for realization and is suitable for integration
within Wireless system.
The simulated filter structure and response is
shown in fig. 1, fig.2 and show the current distribution
REFERENCES
shown in fig 3. In the response graph gain (dB) is
plotted on the Y-axis and frequency (GHz) on the [1] G. Mathaei, L.Y. oung, E.M.T. Jones, “Microwave Filter
impedance matching networks and coupling Structures”, Artech
X-axis. It is clear from fig.2 that the simulation cut off House, Norwood, MA, 1980.
frequency is found is 2.24GHz and insertion loss at [2] R.E. Collin,” Foundation for Microwave Engineering”,Second
3.40GHz is-20.06dB. So the designed stepped Edition, McGraw-Hill, N. Y.1992.
impedance low pass filter is capable if passing [3] M. Makimoto, S.Yamoshita, “Microwave Resonators and
Filters”, IEEE Trans. Wireless Communication Vol. 2, August
frequency less than 2.24GHz and Rejects the frequency 1986.
after 2.24GHz. The return loss performance up [4] Tae-Yeoul Yun, Kai Chang, “Transaction on Microwave Theory
to-37.672dB at 1.27GHz. So this filter can be used for and Techniques”, IEEE Trans, Vol. 49, No. 3, 2001.
[5] Sudipta Das, Dr. S.K.Chowdhury,Senior Member of IEEE,
L-band applications. “Design Simulation and Fabrication of Stepped Impedance
Microstripline Low Pass Filter for S-band Application using
VI. CONCLUSION IE3D and MATLAB” international Journal of Electronics &
Communication Technologies, Vol. 3, Issue 1,PP.98-100, Jan –
A microstrip stepped impedance low pass filter has March 2012.
[6] Navita Singh, Saurabh Dhiman, Prerna Jain, Tanmay Bhardwaj,
been designed and simulated. The low pass filter design “Design of Stepped- Impedance Microstrip line Low Pass Filter
for cut off frequency of 2.5GHz and the simulated filter For Wireless communication”, International Conference on
achieved parameters cut off frequency 2.24 GHz and - Advanced Computing, Communication and Networks’11,
PP. 806–808.
20dB insertion loss at 3.4 GHz. The designed filter can [7] Navita Singh, Dr.Avinash Kumar and Pravesh Singh, “Low Pass
be included at the transmitting and receiving end of Filter for L-band Application using Steppedimpedance
system to get desired spectrum for L-band application. Microstrip Lines”, Journal of Engineering Research and Studies,
E-ISSN 0976–7916.
We have designed stepped impedance low pass filter [8] Jia Shen G. Hong, M. J. Lancaster, “Micro strip Filters for
because it takes less space than a similar low pass filter RF/Microwave Applications”, John Wiley & Sons Inc., 2001.
using stubs. Its compact nature minimizes required [9] D.M. Pozar, “Microwave Engineering”, John Wiley, 2000.
Abstract—An analytical study of one-dimensional However, S.K. Singh et al. obtained large frequency
photonic crystal structure consisting alternate layers TiO2 range of omni-directional reflection by overlapping two
and SiO2 has been presented. Using transfer matrix
photonic crystals [6]. In another investigation, Villar et
method, the reflectivity of TE and TM modes has been
calculated theoretically. It is found that at 1550nm al. have analyzed one-dimensional photonic band gap
wavelength and for an angle of incidence of 780, the structures with a defect layer of liquid crystal for the
reflectivity of the TE mode for such a structure is unity development of fiber-optic tunable wavelength filters
and that of the TM mode is almost zero. This property [7]. Recently, Kumar et al. proposed a simple design of
may be exploited for the design of optical TE mode filter a cascaded photonic band gap filter in ultraviolet region
or TM mode filter. [8]. Qiao et al. [9] reported a new kind of photonic
Kwywords: Photonic crystal, Optical filter, Transfer crystal constructed by different positive index materials
matrix method. is called Photonic Quantum Well (PQW). With the help
of this structure photonic band gaps may be enlarged
I. INTRODUCTION and narrow multichannel filters can be obtained
Many studies on periodic structures for their optical [10-11].
properties have been worked out by many investigators In the present paper we use alternate layers of TiO2
in the last century. The concept of photonic crystals was and SiO2 .One of the most important properties of fused
silica is its extremely low coefficient of expansion: 5.5
put forward in the latter half of the 1980s. Notably, the
x 10-7 mm This makes the material particularly useful
pioneering works of by Yablonovitch [1] and John [2]
for optical flats, mirrors, furnace windows and critical
may be considered as the new beginning of this exciting
optical applications On the other hand, titanium dioxide
field of study. Generally, photonic crystal structures (TiO2) is attracting wide attentions due to its novel
may be considered to be regular arrangements of applications in photo catalysts, sensors and solar cells. .
dielectric materials which may lead to the formation of First of all, TiO2 has a fairly large refractive index
an energy band structure for electromagnetic wave which is large enough to form a photonic band gap in
propagating through them. The majority of applications photonic structures. Secondly, the optical absorption
of photonic crystals exploit the phenomenon of loss of TiO2 is about 10 times lower than that of silicon
photonic band-gap (PBG). Photonic band-gap may be at the optical communication wavelength of 1.5µm.
defined as a frequency region characterized by zero thirdly; its thermal expansion coefficient is very small.
density of the electromagnetic states. The propagation
of electromagnetic waves inside of a photonic crystal is,
therefore, suppressed for the frequencies within PBG.
Because of the existence of PBGs in certain photonic
crystals, the design of efficient low-loss dielectric
reflectors that can confine radiation in channels
(waveguides) or localized defects (resonators) with
sizes comparable to the wavelength of light has become
a possibility.
The possibility of the design of optical band-pass
Fig. 1: Periodic Variation of One Dimensional
filter in near and far infra-red regions based on one Photonic band Gap Structure
dimensional photonic crystals was suggested in 1992 by
Ojha et al [3]. Later on Chen et al. calculated important II. THEORETICAL ANALYSIS
results on photonic filters by using photonic air bridges
[4]. Photonic band gap filter for wavelength division Using transfer matrix method, which is widely used
multiplexing was fabricated by D’Orazio et al. [5]. for the description of the properties of stacked layers
used it to determine the reflectivity and transitivity of 16 sub-layers of AB unit cell. For AB stack, we choose
photonic structures. the material of layer A as TiO2 and the material of layer
Let us consider the stack of N layers perpendicular B as SiO2 having refractive indices 2.65 and 1.45
on the OZ axis as it can be seen in fig.1. The index of respectively. The incident and emergent medium is
refraction profile in the form of rectangular symmetry is taken as air .The thickness of two layers are taken as
given by: a=465 nm and b=255 nm respectively.
⎧⎪n , 0 < z < d1
n( z ) = ⎨ 1 (1)
⎪⎩n2 , d1 < z < d 2
Where n( z ) = n( z + md ) and m = 0, ±1, ±2,.....
and d=d1+d2 is the lattice period.
To determine the electric field, we solve the
following system and find the relation between A0, B0,
As, Bs
⎛ A0 ⎞ ⎛ M 11 M 12 ⎞ ⎛ As ⎞ (2)
⎜⎜ ⎟⎟ = ⎜
⎝ M 21 M 22 ⎟⎠ ⎜⎝ B s ⎟⎠
⎝ 0⎠
B
With the matrix given by
N
⎛ M11 M12 ⎞ −1 ⎡ 2 −1 ⎤ Fig. 2: Dispersion Spectra of TiO2 /SiO2 One Dimensional PC
⎜M ⎟ = D0 ⎢i∏ D PD Ds
⎝ 21 M 22 ⎠ ⎣ =1 i i i ⎥⎦ for TE and TM Polarization with Angle of Incidence at 780
REFERENCES
[1] E. Yablonovitch, “Inhibited Spontaneous Emission in
Solid-State Physics and Electronics”, Phys. Rev. Lett. 58,
pp. 2059–2062, May 1987.
[2] S. John, “Strong localization of photon in certain disordered
dielectric superlattice”, Phys. Rev. Lett. 58, pp. 2486-2489,
May 1987.
[3] S.P. Ojha, P.K. Chaudhary, P. Khastgir and O.N. Singh,
“Operating characteristics of an optical filter with a linearly
periodic refractive index pattern in the filter material”, Jpn. J.
Appl. Phys. 31, pp. 281–285, 1992.
Fig. 5: Reflectance Spectra at Oblique Angle of Incidence780 of [4] J.C. Chen, A. Haus, S. Fan, P.R. Villeneuve and J.D.
TiO2/SiO2 One Dimensional PC for TE and TM Polarizations Joannopoulos, “Optical filters from photonic band gap
air bridges”, IEEE: J. Lightwave Technol. 14, pp. 2575–2580,
Fig. 3, 4 shows the reflectivity and transmittivity of 1996.
TE and TM mode at different angle of incidence. Fig. 5 [5] D’Orazio, M. De Sario, V. Petruzzelli and F. Prudenzano,
shows the reflectivity at different wavelength. On “Photonic band gap filter for wavelength division multiplexer”,
Opt. Exp. 11, pp. 230–239, 2003.
studying this figure it is noticeable that at particular [6] S.K. Singh, Khem B. Thapa and S.P. Ojha, “Large Frequency
angle and wavelength TE and TM mode exhibit unique Range of Omnidirectional Reflection in Si-based One-
properties. At angle 780 and λ=1.55µm which is the Dimensional Photonic Crystals”, Int. J. Micro. Opt. Technol.
1(2), pp. 686–690, 2006.
mean wavelength of optical window and can be used in [7] I. DelVillar, I.R. Matias, F.J. Arregui and R.O. Claus, “Analysis
fiber optics communication, the reflectivity of TE mode of one dimensional photonic band gap structures with a liquid
is unity and its transmittivity is zero while TM mode crystal defect toward development of fiber-optic tunable
shows vice versa i.e. TE waves are completely reflected wavelength filters”, Opt. Exp. 11(5), pp.430–436, 2003.
[8] Vipin Kumar, Kh. S. Singh and S. P. Ojha, “A Simple Design of
and do not pass through the crystal but TM waves are a Monochromator by cascading photonic band gap Filters”, Int.
completely transmitted at this wavelength and angle. J. Micro. Opt. Technol. 5 (3), pp. 162–165, 2010.
[9] F. Qiao, C. Zhang, J. Wan and J. Zi, “Photonic quantum-
wellstructures: Multiple channeled filtering phenomena”, Appl.
IV. CONCLUSION Phys. Lett. 77, pp. 3698–3700, 2000.
[10] Y. Xiang, X. Dai, S. Wen and D. Fan, “Omnidirectional and
It can be concluded that it is possible to achieve TE multiple-channeled high-quality filters of photonic
and TM mode filter by using this property. It is also heterostructures containing single-negative materials”, J. Opt.
called as a demultiplexer because by using that type of Soc. Am. A 24, pp. A28–A32, 2007.
photonic crystal we can increase the storage capacity of [11] Y. H. Chen, “Frequency response of resonance modes in
heterostructures composed of single-negative materials”, J. Opt.
line in fibre optics communication. So we can use the Soc. Am. B 25, pp. 1794–1799, 2008.
proposed device as a tunable demultiplexer in optical
Nanorobotics
K.B. Mohd. Umar Ansari1 and Goldy Sharma2
M.Tech (Electrical Power & Energy Systems), Dept. of Electrical & Electronics Engineering,
Ajay Kumar Garg Engineering College, Ghaziabad, India
e-mail: 1mumar28@gmail.com, 2goldysharmaacet@gmail.com
Abstract—Nanorobotics is the technology of creating word nanorobot is the correct technical term in the
machines or robots at or close to the microscopic scale of a nonfictional context of serious engineering studies.
nanometre. More specifically, nanorobotics refers to the Some proponents of nanorobotics, in reaction to
still largely hypothetical nanotechnology engineering the grey goo scare scenarios that they earlier helped to
discipline of designing and building nanorobots.
Nanorobots would be typically devices ranging in size
propagate, hold the view that nanorobots capable of
from 0.1-10 micrometers and constructed of nanoscale or replication outside of a restricted factory environment
molecular components. As no artificial non-biological do not form a necessary part of a purported productive
nanorobots have yet been created, they remain a nanotechnology, and that the process of self-
hypothetical concept. replication, if it were ever to be developed, could be
Another definition sometimes used is a robot which made inherently safe. They further assert that free-
allows precision interactions with nanoscale objects, or foraging replicators are in fact absent from their current
can manipulate with nanoscale resolution. Following this plans for developing and using molecular
definition even a large apparatus such as an atomic force manufacturing.
microscope can be considered a nanorobotic instrument
when configured to perform nanomanipulation. Also
In such plans, future medical nanotechnology has
macroscale robots or microrobots which can move with been posited to employ nanorobots injected into the
nanoscale precision can also be considered nanorobots patient to perform treatment on a cellular level. Such
nanorobots intended for use in medicine are posited to
I. INTRODUCTION be non-replicating, as replication would needlessly
increase device complexity, reduce reliability, and
Nanomachines are largely in the research-and- interfere with the medical mission. Instead, medical
development phase, but some primitive molecular nanorobots are posited to be manufactured in
machines have been tested. An example is a sensor hypothetical, carefully controlled nanofactories in
having a switch approximately 1.5 nanometers across, which nanoscale machines would be solidly integrated
capable of counting specific molecules in a chemical into a supposed desktop-scale machine that would build
sample. The first useful applications of nanomachines, macroscopic products
if such are ever built, might be in medical technology, The most detailed discussions of nanorobotics,
where they might be used to identify cancer cells and including specific design issues such as sensing, power
destroy them. Another potential application is the communication, navigation, manipulation, locomotion,
detection of toxic chemicals, and the measurement of and onboard computation, have been presented in the
their concentrations, in the environment. Recently, Rice medical context of nanomedicine by Robert Freitas.
University has demonstrated a single-molecule car Although much of these discussions remain at the level
which is developed by a chemical process and includes of unbuildable generality and do not approach the level
buckyballs for wheels. It is actuated by controlling the of detailed engineering, the Nanofactory Collaboration,
environmental temperature and by positioning a founded by Robert Freitas and Ralph Merkle in 2000,
scanning tunneling microscope tip. is a focused ongoing effort involving 23 researchers
from 10 organizations and 4 countries that is
II. NANOROBOTICS developing a practical research agenda specifically
Since nanorobots would be microscopic in size, it aimed at developing positionally-controlled diamond
would probably be necessary for very large numbers of mechanosynthesis and a diamondoid nanofactory that
them to work together to perform microscopic and would be capable of building diamondoid medical
macroscopic tasks. These nanorobot swarms, both nanorobots.
those which are incapable of replication and those
which are capable of unconstrained replication in the III. NUBOT
natural environment, are found in many science fiction Nubot is an abbreviation for Nucleic Acid
stories, such as the Borg nanoprobes in Star Trek. The Robots.Nubots are synthetic robotics devices at the
word nanobotis often used to indicate this fictional nanoscale. Representative nubots include the several
context and is an informal or even pejorative term to DNA walkers reported by Ned Seeman's group at
refer to the engineering concept of nanorobots. The
NYU, Niles Pierce's group at Caltech, John Reif's any introduction of new technology, including concerns
group at Duke University, Chengde Mao's group at about the toxicity and environmental impact of
Purdue, and Andrew Turberfield's group at the nanomaterials,and their potential effects on global
University of Oxford. economics, as well as speculation about various
doomsday scenarios. These concerns have led to a
IV. NANOBOTS IN FICTION debate among advocacy groups and governments on
whether special regulation of nanotechnology is
Nanobots have been a recurring theme in many warranted.
science-fiction novels, sci-fi shows and movies, such as
the sci-fi show Red Dwarf where they are used to A. Fundamental Concepts
manufacture a new arm for Dave Lister from his excess
body tissue. The popular video game series Metal Gear One nanometer (nm) is one billionth, or 10-9, of a
Solid has "nanite" factories in-game. They are used as a meter. By comparison, typical carbon-carbon bond
healing mechanism in the Sony PlayStation series lengths, or the spacing between these atoms in a
Ratchet and Clank. In the game System Shock 2, molecule, are in the range 0.12-0.15 nm, and a DNA
"nanites" are used as currency as well as a type of double-helix has a diameter around 2 nm. On the other
weapon ammo. Nanobots were also featured during the hand, the smallest cellular lifeforms, the bacteria of the
Sci-Fi Channel era of Mystery Science Theater 3000, genus Mycoplasma, are around 200 nm in length.
where they were known as "nanites." They were To put that scale in another context, the
depicted on the show as microscopic, bug-like, comparative size of a nanometer to a meter is the same
freestanding robots with distinct personalities, as that of a marble to the size of the earth. Or another
including a hairdresser and union foreman. In way of putting it: a nanometer is the amount a man's
Stargate: Atlantis there is a race of machines called the beard grows in the time it takes him to raise the razor to
"Asurans" or "Replicators" whose bodies are entirely his face.
built of nanites. The first person shooter Crysis Two main approaches are used in nanotechnology.
includes a suit that consists of nanorobots, which In the bottom-up approach, materials and devices are
replenish armor, health, enhance strength, speed and built from molecular components which assemble
even enable active camouflage. themselves chemically by principles of molecular
recognition. In the top-down approach, nano-objects
A. Potential Applications are constructed from larger entities without atomic-
level control.
• Early diagnosis and targeted drug delivery for
cancer VI. SPECULATIVE
• Biomedical instrumentation
• Surgery Molecular nanotechnology is a proposed approach
• Pharmacokinetics which involves manipulating single molecules in finely
• Monitoring of diabetes controlled, deterministic ways. This is more theoretical
than the other subfields and is beyond current
V. NANOTECHNOLOGY capabilities.
Due to the popularity and media exposure of the
Nanotechnology, which is sometimes shortened to term nanotechnology, the words picotechnology and
"Nanotech", refers to a field whose theme is the control femtotechnology have been coined in analogy to it,
of matter on an atomic and molecular scale. Generally although these are only used rarely and informally.
nanotechnology deals with structures of the size 100
nanometers or smaller, and involves developing A. Tools and Techniques
materials or devices within that size. Typical AFM setup. A microfabricated cantilever
Nanotechnology is extremely diverse, ranging with a sharp tip is deflected by features on a sample
from novel extensions of conventional device physics, surface, much like in a phonograph but on a much
to completely new approaches based upon molecular smaller scale. A laser beam reflects off the backside of
self-assembly, to developing new materials with the cantilever into a set of photodetectors, allowing the
dimensions on the nanoscale, even to speculation on deflection to be measured and assembled into an image
whether we can directly control matter on the atomic of the surface.
scale. The first observations and size measurements of
There has been much debate on the future of nano-particles were made during the first decade of the
implications of nanotechnology. Nanotechnology has 20th century. They are mostly associated with the name
the potential to create many new materials and devices of Zsigmondy who made detailed studies of gold sols
with wide-ranging applications, such as in medicine, and other nanomaterials with sizes down to 10 nm and
electronics, and energy production. On the other hand, less. He published a book in 1914. He used
nanotechnology raises many of the same issues as with
ultramicroscope that employs a dark field method for nanotechnology, bionanotechnology, and nanomedicine
seeing particles with sizes much less than light are used to describe this hybrid field. Functionalities
wavelength. can be added to nanomaterials by interfacing them with
Next group of nanotechnological techniques biological molecules or structures. The size of
include those used for fabrication of nanowires, those nanomaterials is similar to that of most biological
used in semiconductor fabrication such as deep molecules and structures; therefore, nanomaterials can
ultraviolet lithography, electron beam lithography, be useful for both in vivo and in vitro biomedical
focused ion beam machining, nanoimprint lithography, research and applications. Thus far, the integration of
atomic layer deposition, and molecular vapor nanomaterials with biology has led to the development
deposition, and further including molecular self- of diagnostic devices, contrast agents, analytical tools,
assembly techniques such as those employing di-block physical therapy applications, and drug delivery
copolymers. vehicles.
There are several important modern developments.
The atomic force microscope (AFM) and the Scanning VIII. DIAGNOSTICS
Tunneling Microscope (STM) are two early versions of
scanning probes that launched nanotechnology. Various Nanotechnology-on-a-chip is one more dimension
techniques of nanolithography such as dip pen of lab-on-a-chip technology. Biological tests measuring
nanolithography, electron beam lithography or the presence or activity of selected substances become
nanoimprint lithography were also developed. quicker, more sensitive and more flexible when certain
Lithography is a top-down fabrication technique where nanoscale particles are put to work as tags or labels.
a bulk material is reduced in size to nanoscale pattern. Magnetic nanoparticles, bound to a suitable antibody,
are used to label specific molecules, structures or
B. Applications
microorganisms. Gold nanoparticles tagged with short
Most applications are limited to the use of "first segments of DNA can be used for detection of genetic
generation" passive nanomaterials which includes sequence in a sample. Multicolor optical coding for
titanium dioxide in sunscreen, cosmetics and some food biological assays has been achieved by embedding
products; Carbon allotropes used to produce gecko different-sized quantum dots into polymeric
tape; silver in food packaging, clothing, disinfectants
microbeads. Nanopore technology for analysis of
and household appliances; zinc oxide in sunscreens and
nucleic acids converts strings of nucleotides directly
cosmetics, surface coatings, paints and outdoor
furniture varnishes; and cerium oxide as a fuel catalyst. into electronic signatures.
Nano-membranes have been produced that are
portable and easily-cleaned systems that purify, IX. DRUG DELIVERY
detoxify and desalinate water meaning that third-world The overall drug consumption and side-effects can
countries could get clean water, solving many water
be lowered significantly by depositing the active agent
related health issues.
in the morbid region only and in no higher dose than
VII. HEALTH AND ENVIRONMENTAL CONCERNS needed. This highly selective approach reduces costs
and human suffering. Nanotechnology is also opening
Some of the recently developed nanoparticle up new opportunities in implantable delivery systems,
products may have unintended consequences. which are often preferable to the use of injectable
Researchers have discovered that silver nanoparticles drugs, because the latter frequently display first-order
used in socks to reduce foot odor are being released in
kinetics This rapid rise may cause difficulties with
the wash with possible negative consequences. A study
at the University of Rochester found that when rats toxicity, and drug efficacy can diminish as the drug
breathed in nanoparticles, the particles settled in the concentration falls below the targeted range.
brain and lungs, which led to significant increases in
biomarkers for inflammation and stress response. X. TISSUE ENGINEERING
A major study published more recently in Nature Nanotechnology can help to reproduce or to repair
nanotechnology suggests some forms of carbon damaged tissue. This so called “tissue engineering”
nanotubes a poster child for the nanotechnology
makes use of artificially stimulated cell proliferation by
revolution could be as harmful as asbestos if inhaled in
sufficient quantities. using suitable nanomaterial-based scaffolds and growth
factors. Tissue engineering might replace today’s
A. Medicine conventional treatments like organ transplants or
The biological and medical research communities artificial implants. On the other hand, tissue
have exploited the unique properties of nanomaterials engineering is closely related to the ethical debate on
for various applications.Terms such as biomedical human stem cells and its ethical implications.
XI. DNA NANOTECHNOLOGY strands cross from one duplex into the other. This
molecule has the advantage that the junction points are
DNA nanotechnology has been applied to the
now constrained to a single orientation as opposed to
related field of DNA computing. A DX array has been
being flexible as in the four-arm junction. This makes
demonstrated whose assembly encodes an XOR
the DX motif suitible as a structural building block for
operation, which allows the DNA array to implement a
larger DNA complexes.
cellular automaton which generates a fractal called the
Sierpinski gasket. DNA computing can be done without
XIII. DNA NANOTUBES
using the types of molecules made possible by DNA
Nanotechnology. In addition to flat sheets, DX arrays have been
DNA nanotechnology is a subfield of made to form hollow tubes of 4-20 nm diameter. These
nanotechnology which seeks to use the unique DNA nanotubes are somewhat similar in size and shape
molecular recognition properties of DNA and other to carbon nanotubes, but the carbon nanotubes are
nucleic acids to create novel, controllable structures out stronger and better conductors, whereas the DNA
of DNA. The DNA is thus used as a structural material nanotubes are more easily modified and connected to
rather than as a carrier of genetic information, making other structures.
it an example of bionanotechnology. This has possible
applications in molecular self-assembly and in DNA A. DNA Nanomechanical Devices
computing. DNA complexes have been made which change
their conformation upon some stimulus. These are
XII. DNA CROSSOVER MOLECULES intended to have applications in nanorobotics. One of
A double-crossover (DX) molecule. This molecule the first such devices, called "molecular tweezers,"
consists of five DNA single strands which form two changes from an open to a closed state based upon the
antiparallel double-helical domains, on the left and the presence of control strands.
right in this image. There are two crossover points DNA machines have also been made which show a
where the strands cross from one domain into the other. twisting motion. One of these makes use of the
DNA nanotechnology makes use of branched transition between the B-DNA and Z-DNA forms to
DNA structures to create DNA complexes with useful respond to a change in buffer conditions. Another relies
properties. DNA is normally a linear molecule, in that on the presence of control strands to switch from a
its axis is unbranched. However, DNA molecules paranemic-crossover (PX) conformation to a double-
containing junctions can also be made. For example, a junction (JX2) conformation.
four-arm junction can be made using four individual
DNA strands which are complementary to each other in REFERENCES
the correct pattern. Due to Watson-Crick base pairing, [1] Drexler, K. Eric (1986). Engines of Creation: The Coming Era
only portions of the strands which are complementary of Nanotechnology.
to each other will attach to each other to form duplex [2] Binnig, G.; Rohrer, H. (1986). "Scanning tunneling
microscopy". IBM Journal of Research and Development 30: 4.
DNA. This four-arm junction is an immobile form of a [3] "Nanoscience and nanotechnologies: opportunities and
Holliday junction. uncertainties". Royal Society and Royal Academy of
Junctions can be used in more complex molecules. Engineering. July 2004.
The most important of these is the "double-crossover" [4] Kevin Rollins (Nems Mems Works, LLC). "Nanobiotechnology
Regulation: A Proposal for Self-Regulation with Limited
or DX motif. Here, two antiparallel DNA duplexes lie Oversight". Volume 6–Issue 2.
next to each other, and share two junction points where [5] Paull, J. & Lyons, K. (2008). "Nanotechnology: The Next
Challenge for Organics".Journal of Organic Systems 3: 3–22.
XIII. CONCLUSION
Digital watermarking is evolved in the area of
research and development. One key research problem
that we still face today is the development of truly
robust, secure and transparent watermarking technique
for different digital media including image, video,
SEA Result audio, text.
Detecting Another problem is the development of semi
Function fragile authentication techniques. The solution of this
technique application result and development of new
result in the field of information Theory, game theory,
Signal processing, Detection and Decision making
theory. Although a lot of progress has already been [2] Frank Y. Shih: Digital watermarking and steganography
fundamentals and techniques. Taylor & Francis, Boca Raton,
made, there still remain many open issues that need FL, USA, 2008.
attention before this area becomes mature [3] Khan, A. and Mirza, A. M. 2007. Genetic perceptual shaping:
Utilizing cover image and conceivable attack information
REFERENCES during watermark embedding. Inf. Fusion 8, 4 (Oct. 2007),
354–365.
[1] Ingemar J. Cox: Digital watermarking and steganography.
Morgan Kaufmann, Burlington, MA, USA, 2008.
Abstract—The power of wireless sensor network details the different wireless sensor network algorithms
technology has provided the capability of developing large implemented in the landslide detection network. The
scale systems for real-time monitoring. This paper wireless sensor test bed for landslide detection is
describes the evolution of a wireless sensor network described in detail in Section VIII. Field deployment,
system for landslide detection in the Idukki district of the
southern state of Kerala, India, a region known for its
its design concerns, and experiences are described in
heavy rainfall, steep slopes, and frequent landslides. The Section IX. Finally we conclude in Section X and in the
deployment and data retrieval or collection from same section future work is also discussed.
geophysical sensors, the design, development and
deployment of WSN, the development of data collection II. RELATED WORK
and data aggregation algorithms needed for the network,
and the network requirements of the deployed landslide The evolution of wireless sensor networks has
detection system, data analysis system etc has been fostered the development of real-time monitoring of
discussed in this paper. critical and emergency applications. Wireless sensor
technology has generated enthusiasm in computer
Keywords: Wireless Sensor Network, Landslide,
Heterogeneous Networks Distributed algorithms scientists to learn and understand other domain areas
which have helped them to propose or develop real-
I. INTRODUCTION time deployments. One of the major areas of focus is
environmental monitoring, detection and prediction.
Real-time monitoring of environmental disasters The Drought Forecast and Alert System (DFAS) has
are one of the prime necessity of the world. Different been proposed and developed in paper [10]. This
technologies have been developed for this purpose. system uses mobile communication to alert the users,
Wireless sensor networks (WSN) is one of the major whereas the deployed system uses real time data
technology that can be used for real-time monitoring. collection and transmission using the wireless sensor
WSN has the capability of large scale deployment, nodes, WiFi, satellite network and also through
low maintenance, scalability, adaptability for different internet. The real streaming of data through broadband
scenarios etc. WSN has its own limitation such as low connectivity provides connectivity to wider audience.
memory, power, bandwidth etc, but its capability to be An experimental soil monitoring network using a
deployed in hostile environment, and low maintenance wireless sensor network is presented in reference [11],
requirement made it one of the best suited technology which explores real-time measurements at temporal and
for real-time monitoring. spatial granularities which were previously impossible.
This paper discusses the design and deployment of This paper also discusses about the reception of real-
a landslide detection system using wireless sensor time measurements at temporal and spatial granularity.
network1 at Anthoniar Colony, Munnar, Idukki (Dist), Research has shown that other than geotechnical
Kerala (State), India. The deployment site has sensor deployment and monitoring, other techniques
historically experienced several landslides, with the such as remote sensing, automated terrestrial surveys,
latest one occurring in the year 2005, which caused a and GPS technology,etc. also can be used by
death toll of 10 (people). themselves or in combination with other technologies
The remainder of the paper is organized as follows. to provide information about land deformations. Paper
Section II describes related work in WSN systems, and [12] describes a state-of-the-art system that combines
other methods for landslide prediction. In Section III, multiple sensor types to provide measurements to
we describe about landslide phenomena and Section IV perform deformation monitoring.
describes about the sensors needed for monitoring Reference [9] discusses the topic of slip surface
rainfall induced landslides. Section 5 details about the localization in wireless sensor networks, which can be
enhanced sensor column design used along with this used for landslide prediction. A durable wireless sensor
system. Section VI describes about the wireless sensor node has been developed [13], which can be employed
architecture used for landslide scenario and Section VII in expandable wireless sensor networks for remote
monitoring of soil conditions in areas conducive to Himalayan belt. High rainfall intensity accelerates the
slope stability failures. In this paper, real time sliding and slumping in the existing hazard zones. The
deployment of a heterogeneous network in India for annual loss due to landslides in India is equivalent to
landslide detection has been discussed. This study $400 million.
incorporates both theoretical and practical knowledge
from diverse domains such as landslides and IV. SENSORS NEEDED FOR MONITORING RAINFALL
geomechanics, wireless sensor, Wi-Fi, and satellite INDUCED LANDSLIDES
networks, power saving solutions, and electronic
Under heavy rainfall conditions, rain infiltration on
interface and design, among others, which paved the
design, development and deployment of a real-time the slope causes instability, a reduction in the factor of
landslide detection system using a wireless sensor safety, transient pore pressure responses, changes in
water table height, a reduction in shear strength which
network.
holds the soil or rock, an increase in soil weight and a
reduction in the angle of repose. When the rainfall
III. LANDSLIDE
intensity is larger than the slope saturated hydraulic
Landslide is a general term used to describe the conductivity, runoff occurs [3]. Three distinct physical
down slope movement of soil, rock and organic events occur during a landslide:
materials under the influence of gravity. It can be • The initial slope failure.
triggered by gradual processes such as weathering, or • The subsequent transport.
by external mechanisms including: • The final deposition of the slide materials.
• Undercutting of a slope by stream erosion, The initial slope failure can occur due to the
wave action, glaciers, or human activity such increase in pore pressure and soil moisture content,
as road building. under heavy rainfall, which necessitates the inclusion
• Intense or prolonged rainfall, rapid snowmelt, of geophysical sensors for detecting the change in pore
or sharp fluctuations in ground-water levels. pressure and moisture content with the warning system
• Shocks or vibrations caused by earthquakes or developed for landslide detection .So the system
construction activity. discussed in this paper also includes geophysical
• Shocks or vibrations caused by earthquakes or sensors such as pore pressure transducer and dielectric
construction activity. moisture sensor for capturing the in-situ measurements.
• Loading on upper slopes. After the slope failure the subsequent transport of
• A combination of these and other factors. the material happens that will generate slope gradient
Once a landslide is triggered, material is change, vibration etc, which has to be measured and
transported by various mechanisms including sliding, monitored for effective issue of warning. So the
flowing and falling.The types of landslides vary with warning system includes strain gauge, and tiltmeter,
respect to the: that can be used for measuring in-situ slope gradient
• Rate of movement: This ranges from a very changes. Along with them geophone is used for
slow creep (millimetres/year) to extremely analysing the vibration.
rapid (metres/second).
• Type of material: Landslides are composed of V. ENHANCED SENSOR COLUMN DESIGN
bedrock, unconsolidated sediment and/or Commercially available wireless sensor nodes do
organic debris. not have implanted sensors to measure pore pressure,
• Nature of movement: The moving debris can moisture content, vibration, earth movements, etc. This
slide, slump, flow or fall. constraint has lead us to implement data acquisition
Landslides constitute a major natural hazard in boards to connect the external sensors to the wireless
India that accounts for considerable loss of life and sensor nodes. The geological sensors were placed
damage to communication routes, human settlements, inside a sensor column and they were connected to the
agricultural fields and forest lands. The Indian wireless sensor node via a data acquisition board as
subcontinent, with diverse physiographic, shown in Figure 1. The sensor column design discussed
seismotectonic and climatological conditions is in this paper is an enhanced version of the sensor
subjected to varying degree of landslide hazards; the column discussed in [9], which uses a homogeneous
Himalayas including Northeastern mountains ranges structure, whereas our design uses a heterogeneous
being the worst affected, followed by a section of structure which differs with respect to the terrain
Western Ghats and the Vindhyas. conditions and the geological and hydrological
In India, landslides mainly happen due to heavy parameters of the deployment site. Also, in this sensor
rainfall, so this study concentrates on rainfall induced column design all the geological sensors (such as
landslides. Earthquakes can also cause landslides, geophone and dielectric moisture sensor) are not placed
however in India this is primarily confined to the inside the column but are connected to the same
wireless sensor node. The sensor column design also and consensus to arrive at reliable decisions, more
includes tilt meters which can be used for validating the accurate than that of each single sensor and capable to
deformation measurements captured using strain achieve globally optimal decisions as discussed in
gauges. research papers [16],[17], and [18].
threshold based algorithm [14]. This algorithm will (gateway) kept at the deployment site. Data received at
help to reduce the energy consumed during the low the gateway has to be transmitted to the Field
alert levels and also in collecting and transmitting large Management Center (FMC) which is approximately
amounts of data, only when the environmental and 500m away from the gateway. A Wi-Fi network is used
geological conditions demand the same. Other than between the gateway and FMC to establish the
these methods, state level transitions have been connection. The FMC incorporates facilities such as a
incorporated to reduce the energy consumption per VSAT (Very Small Aperture Terminal) satellite earth
node which will also contribute to reduced energy station and a broadband network for long distant data
consumption throughout the network. These transmission. The VSAT satellite earth station is used
requirements, however, lead to the need of time for data transmission from the field deployment site to
synchronization, and the algorithm planned for the Data Management Center (DMC), situated at our
implementation in our network is discussed in research university campus 300 km away, while the broadband
paper[19]. connection provides fault tolerance for long distance
transmission and can also be used for uploading real
time data directly to a web page with minimum delay.
The DMC consists of the database server and an
analysis station, which performs data analysis and
landslide modeling and simulation on the field data to
determine the landslide probability.The wireless sensor
network architecture for landslide detection is shown in
Figure 4.
Extensive field investigations were conducted for the network. The other sensor column is attached with
identifying the possible locations for sensor column movement sensors since the location of it is in an
deployment.At the deployment site, Anthoniar Colony, unstable region. This sensor column has three tilt
an initial twenty sensor column locations consisting of meters(1m, 2m, 3.5m), and three strain gauges(1.5m,
150 sensors total, were identified with respect to their 2.5m, 4m ) to capture the earth movement from the
geological relevance. The pilot deployment consists of sensor column bending. A dielectric moisture sensor is
two sensor columns, with ten sensors,are deployed in also connected to the sensor column at 1 feet depth.
the field along with six wireless sensor nodes are The wireless sensor nodes sample these sensors at
shown in Figure. every five minutes and sent the data to upper level
sensor nodes in the network. The spatial granularity
will be increased by further addition of more sensor
columns (approximately 20) and wireless sensor nodes
(approximately 20) which is in process.
A. Design and Deployment of Wireless
Sensor Network
The design and development of a wireless sensor
network for the landslide scenario involves
consideration of different factors such as terrain
structure, vegetation index, climate variation,
accessibility of the area etc. The prerequisites of
wireless sensor network development are selection of
sensor column location, sensor column design and its
data collection method, understanding transmission
range and necessity of external antennas or additional
relay nodes, identification of the communication
protocol, development of application specific
algorithms for data aggregation, routing and fault
tolerance etc. The wireless sensor network architecture
at the deployment is discussed in Section 4, and the
wireless senor nodes used for the deployment are 2.4
GHz MicaZ motes from Crossbow. The initial gateway
was stargate with Intel XScale processor 400MHZ and
running ARM Linux OS was programmed as the Sink
Node, while the new gateway is a single board
computer which has 64 MB RAM, 32 MB flash and a
fixed base mote that is used to send and receive the
messages through the transceiver. The sensor column is
physically attached to a wireless sensor node which is
integrated with a data acquisition board. The distance
between current sensor columns is approximately 50
Fig. 5: Field Deployment
meters, at a slope of about 70ˇr. Due to the terrain
structure and vegetation, the data from the sensor
X. DEPLOYMENT OF SENSOR COLUMN
columns are not able to reach the gateway. The major
One of the sensor columns is deployed at the toe reasons for this is, no line of sight path between the
region where various water seepage lines converge. columns,between the first sensor column and the
This fact lead to the installation of pore pressure gateway, and between the second sensor column and
transducers at different depths (2m, 5m) of the sensor the gateway. The above observations, along with
column 1. The other geophysical sensors attached to experimental tests, have led us to employ three relay
this sensor column are dielectric moisture content nodes in between the sensor columns themselves and
transducer, and a geophone. Both pore pressure the gateway. One of the relay nodes is a cluster head
transducer and the dielectric moisture sensor are for the first and second sensor column. The data from
sampled at every five minutes, where as geophone is the cluster head is transmitted to the gateway in the
sampled at the rate of 10 samples/second. The MicaZ
form of packets. At the gateway the received packets
wireless sensor node connected to the sensor column
are time stamped and stored.
transmits the digitized data values to the upper layers of
A Tutorial Overview
on the DGS to Design Microwave Devices
Vikash Kumar1 and Sanjay Gurjar2
1,2
Bhagwant University, Ajmer
e-mail: vikashvvit@gmail.com, 2ersanjay86@yahoo.in
1
Abstract—A tutorial overview on defected ground After all a new structure was developed that was
structure using microstrip line is developed to design defected ground structure (DGS) firstly proposed by
various types of microwave devices like filter, oscillator, park et al. in 1999. DGS alleviate this problem by
coupler, resonator etc. The Defected Ground plane is the connecting two PBG cells with a thin slot which causes:
etched shape on the backside of the metallic ground plane
induces inductive and capacitive effects of the circuits. The • The circuit area becomes relatively small
induced inductive and capacitive effect due to DGS without periodic structures because only a few
provides better roll-off rate and improved slow wave DGS elements have the similar typical
structure. There are different shape are etched to fulfill properties as the periodic structure like the
our requirement in the field of microwave communication. stop-band characteristic.
In this tutorial overview the frequency characterization • DGS needs less circuit size for only a unit or a
and slow wave structure properties is discussed. The few periodic structures showing slow-wave
improved slow-wave structure provides the circuit size
effect.
reduces up-to the 40%. In this paper discussed how the
inductive and capacitive effect induces on the metallic
DGS has found its application in the design of
defected ground plane and discussed how slow wave planner circuits, low-pass filter, high-pass filters, band-
structure improved and how it can calculate analytically pass filter, resonator, oscillator, couplers, etc.
due to the improved frequency characterization. It DGS have gained significant interests. It rejects
provides better cutoff frequency and attenuation pole in certain frequency bands and hence it is called
same frequency without any periodic array of defected electromagnetic band gap (EBG) structure. Figure (1)
ground structure. gives the schematic of such a DGS with its approximate
Keywords: DGS, roll-off rate, slow-wave structure, surface area.
PBG.
I. INTRODUCTION
Now a day wireless communication is daily need of
our life. So we need to how we make this system
flexible. By which it have better noise immunity, larger
bandwidth with less required area, low power
consumption, high reability, compact effect, high
quality miniature components with a dual-band
operation and so many things.
For this we want a single system like PCB by
which we make our all requirements like smart board or
substrate at which we design our all required circuit that
are used in our wireless communication system.
At first we need to know about photonic band-gap Fig. 1: DGS Unit Cell Diagram
(PBG) structure that is a periodic structure which has
DGS is realized by etching a defective patter in the
been known as providing rejection of certain frequency
ground plane, which disturbs the shield current
band. However, it is difficult to use a PBG structure for
distribution in the ground plane. This disturbance can
the design of the microwave or millimeter wave
change the characteristics of a transmission line such as
component due to the difficulties of modeling.
equivalent capacitance and inductance to obtain the
There are so many design parameters that effect on slow wave effect and band-stop property.
the band gap property, such as the number of lattice A DGS for the microstripline has etched-defects in
shape, lattice spacing and relative volume fraction. a metallic ground plane like photonic band gap
Another problem is caused by the radiation from the structures. Since an etched-defect of the ground plane
periodic etched defects. disturbs the shielded current distribution in the ground
II. FREQUEN
NCY CHARACT
TERISTICS OF DGS
D
The proposedd DGS unit seection can pro ovide cutoff
frequuency and attenuation
a poole in some frequency
with
hout any perriodic array of DGS. The T cutoff
frequuency is mainnly dependent to the etched square area
Fig. 2: Threee-dimensional viiew of the Propossed DGS Unit Secction,
which iss ETCHED in the Ground Plane off a Microstrip Linne in grround plane. There
T is also aattenuation po
ole location,
whicch is due to the etched gap ddistance. The presence of
Whichh give rise to increasiing the effeective a cappacitance facttor should be needed. The etched gap
capacitancee and inducctance of a transmission line area,, which is plaaced under a conductor lin ne shown in
respectivelly. Thus, ann LC equivvalent circuit can figurre (2), provvides the paarallel capacitance with
represent the
t proposed unit
u DGS circcuit. The effeccts of effecctive line indductance. Thhus, the prop posed DGS
physical dimensions
d off the proposeed DGS on these sectiion is fully desscribed by twoo parameters:
equivalent circuit parammeters are desscribed. To design
d • The etcheed dimension
circuits with
w the propposed DGS unit section, the • Gap distaance
equivalent circuit and parameters
p forr the DGS seection
should be extracted. These
T are various methodds by III. EQU
UIVALENT CIR
RCUIT OF DGS
which prooposed DGS unit section derived like field
analysis method.
m The eqquivalent circcuit parameterrs are This is a simlipfied
s ciircuit that reeduced the
extracted based on thhe circuit analysis theoryy. By comp plexity of circcuit. So that aany proposed section to a
employing the extracteed parameterrs and the circuit c practtical circuit design,
d it is nnecessary to extract the
analysis thheory, the bannd gap effectt for the provvided equiv valent circuitt parameters. In order to derived the
DGS sectioon can be explained. equiv valent circuiit parameterss of DGS unit u at the
A vaariety of sloot geometries etched inn the referrence plane, the
t S-parametters vs frequeency should
microstrip line ground plane.
p This is shown
s in figurre (3) be calculated
c by full-wave eleectromagnetic simulation
below: to ex
xplain the cutooff and attenuuation pole chaaracteristics
of th
he DGS section.
120 Bhag
gwant Institute of
o Technology, Muzaffarnagar,
M Uttar Pradesh, India
A Tutorial Overview on the DGS to Design Microwave Devices
Fig. 7: New Design and Analysis Method of DGS: (a) Analysis Method of DGS, (b) Equivalent-circuit Model of Unit Cell DGS
Abstract—There are two key factors involved in this left to dangle while the winning node is connected to all
paper. These factors are to obtain accurate bridging fault the destinations of both nodes.
simulation results and to present a hybrid technique that
improves the speed and accuracy of bridging fault
simulation for gate-level standard cell designs. The results
of both combinational and sequential circuits are
compared after study with several other bridging fault
simulators.
I. INTRODUCTION
This is very essential to drive the defect levels of
integrated circuits to extremely low levels. The
traditional single stuck-at fault model does not always
predict the behavior of physical defects in CMOS
technologies. Fault simulation using only this model is
no longer an accurate estimater of IC quality. It is the
drawback of this model and it indicates that there is a
need for more detailed and realistic fault model and
associated fault simulators.
The fault resulting from fabrication defects, Fig. 1: Example Circuit with Bridge form G to H
modeled as dust particles of various size on photomask,
TABLE 1: GATE INPUT PATTERN IMPACT ON OUTPUT
is the bridging fault.This fault type will be the focus of BRIDGE VOLTAGE
this work. Recent work has shown that bridging faults
cause more complex circuit behaviour. Simulation A[0] A[1] A[2] A[3] B C VGH (V)
0 0 0 0 0 1 4.25
method to deal with these complexities exist, but can be
0 0 0 0 1 1 3.68
computationally expensive or limited to combinational 0 0 0 1 0 1 4.13
circuits. This paper will present a fast bridging fault 0 0 0 1 1 1 3.25
simulation method that accurately models the behaviour 0 0 1 1 0 1 3.68
of bridging faults by considering circuit-level near 0 0 1 1 1 1 2.54
bridging site while using gate-level simulation every 0 1 1 1 0 1 2.54
0 1 1 1 1 1 1.38
where else. 1 1 1 1 0 0 2.66
III. ACCURACY OF BRIDGING FAULT MODELS techniques that can deliver the accuracy of a circuit-
level simulator with the speed of a gate-level simulator.
There are two main factors that differentiate the
accuracy of the models: the determination of the volt-
V. REALIZATIONS AND IDEAS
age value that results at the site of the bridge, and the
interpretation of that voltage value by the receivers (the Appying the mechanics of the various bridging
gates downstream of the bridge). fault simulation methods, these observations are made:
The connection of two nodes via a bridging defect • The vast majority of time spent in bridging
implies that the gates driving those two nodes will be fault simulation is in the digital field of interest
engaged in a drive fight (logic contention) when they and is very much simiilar to single stuck-at
have opposite values. fault simulation,
The second factor that determines the accuracy of a • Circuit-level details need to be used only at the
bridging fault model is interpretation of the voltage at bridging site,
the bridged node by the receiving logic gate. • All likely bridging sites can be identified from
A CMOS digital logic gate is a very high gain lay-Out,
amplifier that causes its output to be forced to a power • The only information needed to model the
supply rail when the sensitized input is forced above or behavior of a bridge correctly is the short-
below its logic threshold. The input logic threshold of a circuit voltage of the two driving gates and the
gate input is defined as the voltage value at which the logic threshold of all receiving gates.
input and output of the gate are equal (assuming all • Only a limited number of short-circuit voltages
other inputs to the gate are held at noncontrolling are possible based on the identified bridging
values). A small deviation of the input voltage above or sites and the number of cells in the library,
below the logic threshold, typically a few tenth of a • All logic threshold voltages are known from
volt, is sufficient to cause a large swing in the output. li-brary characterization.
The decision on how to interpret a bridging voltage • Given these realizations about the nature of the
must be made with respect to the input logic threshold procedures essential to bridging fault
voltage of each input of each gate downstream from the simulation, the following ideas suggest
bridge. themselves:
The 3.7 V output value is well above the logic • Use the highest performance single stuck-at
threshold of any input for any gate in the library, sim-ulator available as the underlying core
making this a solid logic one. This example shows that simulator.
a single node value (the bridged node G H) is • Precompute the short-circuit voltage for all
combinations of gates (and all pertinent input
interpreted in all three possible (different) ways: 0, X,
combi-nations) that can be involved in a
and 1.
bridging fault.
The implication of this behavior is disconcerting:
• Precompute the logic threshold voltages for all
different branches from a single fanout stem can
library cell inputs.
effectively have different logic values! Typical gate and
• Modify the fault simulator's evaluation routine
switch-level simulators use a single value to represent
to specially treat the fault site by using the
the voltage on each node (stem) in the circuit and thus
tabu-larized short-circuit voltages and logic
cannot recognize this behavior without special threshold voltages.
modifications. Even then, the restriction to digital The drawback of the new technique is that a single
values (0, 1, and X) eliminates the possibility of stuck-at fault simulator needs to be extended to per
performing comparisons between bridging voltages and form one additional type of evaluation: determining the
input logic thresholds. These restrictions prevent the digital values at gate inputs that are immediately
accurate interpretation of bridging voltages by gate and downstream from the fault site. This is accomplished in
switch- level bridging fault simulators. two phases:
1. Use the bridging defect's driving gates and
IV. FAST AND ACCURATE BRIDG-ING their input values to index into a precomputed
FAULT SIMULATOR short- circuit voltage table to obtain the voltage
The tradeoff between speed and accuracy in seen by the receiving gates, then
bridging fault simulation techniques should be quite 2. Compare that voltage to the precomputed logic
clear by now. Only circuit-level techniques have threshold voltage for each fanout of the bridge
sufficient in- formation to produce correct results, while site to obtain a digital value for simulation.
only gate- level techniques can perform digital
simulation with sufficiently high throughput. This VI. DIFFERENCES FROM PREVIOUS WORK
section presents a hybrid approach based on these two
• The use of explicit SPICE-derived data for
every gate input is unique to this simulator.
This feature is key to obtaining accurate of gate types and the number of gate fanins. If the
simulation results. number of cells in the library is very large, or if the
• Earlier simulators were also limited to number of unique ways to drive ones and zeros is very
combinational or fully scanned circuits, large, the lookup table for the bridging volt- ages can
while this work handles sequential circuits. become unwieldy. The gate input threshold voltage
table grows only linearly with the number of gate types
VII. ASSUMPTIONS and gate fanins in the library and is not a limiting factor.
Another source of error in the precomputed data is
The following assumptions were made in the devel- an incorrect assumption for the value of the likely
opment of the simulator and are responsible for both its bridging resistance. If the resistance of the actual
speed and its limitations: bridging defect is significantly different from that
• Circuits will be standard cell ASICs. as- sumed during characterization, the simulator can
• Only routing channel bridging faults will be pre- dict incorrect results. By repeating the SPICE
considered. char- acterization of the short-circuit voltages with a
• The circuits will be represented at the logic new value of the bridging resistance, a more accurate
gate level. table can be created for the simulator to use.
• The size of the standard cell library is
reasonably small (100 gate types or less). X. CONCLUSION
The practical import of these assumptions on the
simulator is: For prediction of the behavior of a bridging fault
• Only logic gate output shorts need to be correctly, there are two key steps to model physical
handled. mechanisms that must be taken. (i) determining the
• The storage space for and cost of library voltage at the bridging site by resolving the drive fight
characterization is acceptable. between the bridged gates, and (ii)interpreting that
voltage by comparing it to the logic threshold voltages
of each gate input connected to the bridged node. The
VIII. SIMULATION ALGORITHM
BRIDGESIM program uses precomputed values for
In order to maximize fault simulation performance, these voltages to obtain accurate values to feed into a
the differential-based PROOFS algorithm was chosen to gate-level simulation platform. This technique delivers
be the platform upon which the bridging fault simulator an excellent throughput and accuracy, but is limited by
would be built. The basic algorithm had to be extended the scalability of the table lookup procedure and the
to perform the following functions: assumptions made during circuit-level characterization.
• Inject bridging faults instead of stuck-at faults,
• Evaluate the short-circuit voltage at a bridging REFERENCES
site, [1] A. Pancholy, J. Rajski, and L.J. McNaughton, "Empirical failure
• Compare the short-circuit voltage with the analysis and validation of fault models in CMOS," in
logic threshold voltage of fanout gates to Proceedings of the IEEE International Test Conference,
pp. 938-947, Sept. 1990.
deduce new logical values, [2] F. Ferguson and J. Shen, "Extraction and sim- ulation of realistic
• Resolve feedback bridging faults. CMOS faults using inductive fault analysis," in Proceedings of
the IEEE In-ternational Test Conference, pp. 475-484,
IX. LIMITATIONS Sept. 1988.
[3] K.C.Y. Mei, "Bridging and stuck-at faults," IEEE Transactions
The use of circuits that are completely represented on Computers, pp. 720-727, July 1974.
at the gate level and whose bridging faults exist only [4] J.M. Acken and S.D. Millman, "Accurate mod- eling and
simulation of bridging faults," in Pro- ceedings of the IEEE
between gate outputs restricts the applicability of this Custom Integrated Circuits Conference, pp. 17.4.1-17.4.4,1991.
simulator. There is no guarantee that all likely bridg- [5] J.M. Acken and S.D. Millman,"Fault model evolution for
ing faults on a any chip will occur only between gate diagnosis: Accuracy vs. precision," in Proceedings of the IEEE
outputs (intragate faults being the chief example). Table Custom Integrated Circuits Conference, pp. 13.4.1-13.4.4,1992.
[6] G.S. Greenstein and J.H. Patel, "E-PROOFS: A CMOS bridging
3 clearly shows that only a minority (between a quarter fault simulator," in Pro-ceedings of the IEEE International
and a third) of actual bridging faults are gate output Conference on Computer-Aided Design, pp. 268-271, Nov.1992
shorts. It should be noted, however, that shorts between [7] C. Timoc, M. Buehler, T. Griswold, C. Pina, F. Scott, and L.
gate outputs tend to be the most probable type of bridge Hess, "Logical models of physi- cal failures," in Proceedings of
the IEEE Interna- tional Test Conference, pp. 546-553,
because these signals are usu- ally routed over relatively Oct. 1983.
large areas and are thus prone to spot defects. [8] R.E. Bryant, "COSMOS: A compiled sim- ulator for MOS
The scalability of this simulator can be limited by circuits," in Proceedings of the ACM/IEEE Design Automation
the size of the lookup table for short-circuit voltages, Conference, pp. 9-16, June 1987.
which can grow geometrically with both the num- ber
Abstract—During the last few decades, mobile MHZ band and each channel have a one way bandwidth
communication has developed rapidly. The increasing of 30 KHZ for each duplex channel for the U.S. AMPS.
dependency of people on telecommunication resources is B. Second Generation (2G)
pushing even more current technological developments in
the mobile world. Mobile networks was superseded in the 1990s,
1G mobile networks being a wholly analogue (GSM, initially Group Special Mobile later changing
network, provide voice calls. This was superseded by 2G. name to Global System for Mobile Communications)
It introduced services like text messaging (SMS) and which being digital along with voice it introduced
email, along with voice calls. With 2.5G or enhanced 2G, everyone to text messaging (SMS, Short Message
internet access became popular which was earlier limited
only to WAP. 3G mobile telecommunication is a
Service). 2G technologies can be divided into TDMA-
generation of standards for mobile phones and mobile based and CDMA based standards depending on the
telecommunication services fulfilling the International type of multiplexing used [2,6]. The main 2G standards
Mobile Telecommunications-2000 (IMT-2000) are:
specifications by the International telecommunication 1. GSM (TDMA-based), originally from Europe
Union[1,2]. It promises to deliver higher bandwidth and but used in almost all countries on all six
ubiquitous wireless access. 4G technology prescribes that inhabited continents. Today, it accounts for
users can enjoy mobility, seamless access and high quality over 80% of all subscribers around the world.
of service in an all-IP network. 4G will be the certain path
Over 60 GSM operators are also using
to the future radio and mobile communication system[8].
This paper presents the drawbacks of 3G technology
CDMA2000 in the 450 MHz frequency band
and advance features of 4G technology. And, it also (CDMA450).
describe framework of 4G technology. Apart from this, we 2. IS-95 aka cdmaOne (CDMA-based, commonly
will discuss core technologies of 4G mobile system referred as simply CDMA in the US), used in
including security strategy and adaptive network in 4G. the Americas and parts of Asia. Today, it
Keywords: GSM, 2G, 3G, 4G, CDMA, TDMA
accounts for about 17% of all subscribers
globally. Over a dozen CDMA operators have
migrated to GSM.
I. INTRODUCTION
3. PDC (TDMA-based), used exclusively in
A. First Generation (IG) Japan.
4. iDEN (TDMA-based), proprietary network
Mobile networks were deployed in late 1970’s and used by Nextel in the US and Telus Mobility
early 1980's, being a wholly analogue network, in Canada.
providing voice calls. Some of the most successful 1G 5. IS-136 aka D-AMPS (TDMA-based,
system is Nordic Mobile Telephone (NMT) System and commonly referred as simply 'TDMA' in the
Advance Mobile Phone System (AMPS)[3]. US), was once prevalent in the Americas but
1) Nordic Mobile Telephone (NMT) System most have migrated to GSM.
This system was developed in Europe in 1981. The C. Second and a Half Generation (2.5G)
two variants of NMT are: NMT-450 and NMT-900. Is used to describe 2G-systems that have
The numbers indicate the frequency bands uses. NMT- implemented a packet-switched domain in addition to
900 was introduced in 1986 because it carries more the circuit-switched domain. It does not necessarily
channels than the previous NMT-450 network. provide faster services because bundling of timeslots is
2) Advance Mobile Phone System (AMPS) used for circuit-switched data services (HSCSD) as
well.
This system was developed by U.S. Federal The first major step in the evolution of GSM
Communications commission in 1983 and allocated 666 networks to 3G occurred with the introduction of
duplex channels with 40 MHZ of spectrum in the 800 General Packet Radio Service (GPRS). CDMA2000
networks similarly evolved through the introduction of
1x1RTT. The combination of these capabilities came to organization specified the IMT-Advanced (International
be known as 2.5G. Mobile Telecommunications Advanced) requirements
for 4G standards, setting peak speed requirements for
D. Third Generation (3G) 4G service at 100 Mbit/s for high mobility
Mobile telecommunication is a generation of communication (such as from trains and cars) and 1
standards for mobile phones and mobile Gbit/s for low mobility communication (such as
telecommunication services fulfill the International pedestrians and stationary users)[10,11].
Mobile Telecommunications-2000 (IMT-2000) A 4G system is expected to provide a
specifications by the International telecommunication comprehensive and secure all-IP based mobile
Union. Application services include wide-area wireless broadband solution to laptop, computer, wireless
voice telephone, mobile Internet access, video calls and modems, smartphones, and other mobile devices.
mobile TV, all in a mobile environment [3,4]. Facilities such as ultra-broadband Internet access, IP
To meet the IMT-2000 standards, a system is telephony, gaming services, and streamed multimedia
required to provide peak data rates of at least 200 kbit/s may be provided to users.
(about 0.2 Mbit/s). IMT-Advanced compliant versions of LTE and
The following standards are typically branded 3G: WiMAX are under development and called "LTE-
1. The UMTS system, first offered in 2001, Advanced" and "Wireless MAN-Advanced"
standardized by 3GPP, used primarily in respectively. ITU has decided that LTE Advanced and
Europe, Japan, China (however with a Wireless MAN-Advanced should be accorded the
different radio interface) and other regions official designation of IMT-Advanced. On December 6,
predominated by GSM 2G system 2010, ITU recognized that current versions of LTE,
infrastructure. The cell phones are typically WiMAX and other evolved 3G technologies that do not
UMTS and GSM hybrids. Several radio fulfill "IMT-Advanced" requirements could
interfaces are offered, sharing the same nevertheless be considered "4G", provided they
infrastructure: represent forerunners to IMT-Advanced and "a
2. The original and most widespread radio substantial level of improvement in performance and
interface is called W-CDMA. capabilities with respect to the initial third generation
3. The TD-SCDMA radio interface was systems now deployed"HYPERLINK \l "cite_note-
commercialized in 2009 and is only offered in ITUSeminar-."
China.
A. Advanced Features of 4G Technology
4. The latest UMTS release, HSPA+, can provide
peak data rates up to 56 Mbit/s in the downlink
in theory (28 Mbit/s in existing services) and
22 Mbit/s in the uplink.
5. The CDMA2000 system, first offered in 2002,
standardized by 3GPP2, used especially in
North America and South Korea, sharing
infrastructure with the IS-95 2G standard. The
cell phones are typically CDMA2000 and IS-
95 hybrids. The latest release EVDO Rev B
offers peak rates of 14.7 Mbit/s downstream.
E. Limitations of 3G
1. It is difficult to extend to higher data rate in
CDMA.
2. It is difficult to provide a full range of multi-
rate services and not a fully integrated System.
3. There is also a propagation problem in CDMA;
it may not work in a multi path systems from
private to public and indoor to wide area
[9,17].
Fig. 1: Evolution of Wireless Technology
II. 4G TECHNOLOGY • Wider mobile coverage area
In telecommunications, 4G is the fourth generation • Smoother quicker handoff.
of cellular wireless standards. It is a successor to the 3G • Broader bandwidth - higher data rates
and 2G families of standards. In 2009, the ITU-R • WLAN for hot spots, an extension of 2G
and 3G.
A. Protocols & Modulation channel conditions and low or high data rate
users.
Internet Protocol version 6 (IPv6) is considered to 6. Miscellaneous services can be delivered
be the evolved IP protocol for mobile systems through a mixture of transmission networks
providing: including unicast, multicast, and broadcast
1. Enhanced address space network. According to the service types, e.g.\
2. Security real-time attribute, importance, bandwidth
3. Extension headers demand, or data stream type, multiple levels of
B. Session Initiation Protocol (SIP) QoS can be defined for various services.
7. Network resource can be dynamically
It has a flexible structure and can be easily allocated to cope with varying traffic load,
extended. Therefore, it plays a role of true multimedia channel condition, and service environment.
protocol that controls and transports: Traffic conditions will be dynamically
1. Emails 2. Pictures monitored and controlled via techniques such
2. WEB links 4.Videos Service scripts as distributed and decentralized control of
3. Speech 6. Multiparty sessions network.
C. Wideband Code Division Multiple
VII. CONCLUSION
Accesses (WCDMA)
As the history of mobile communications had
It is a wideband direct sequence CDMA system.
presented, legion attempts had been made to reduce a
User information bits are spread over a wide bandwidth
number of technologies to a single global standard.
by multiplexing. The high data bit rates (2Mbps) can be
Projected 4G systems offer promise of a standard that
support by use the variables Spreading factor and
can be embraced worldwide through its key concept of
Multicode Connection. The Most important feature of
integration. Future wireless networks will need to
WCDMA is Power Control, Particularly on uplink[15].
support diverse IP multimedia applications to allow
sharing of resources among multiple users. There must
VI. ADAPTIVE 4G NETWORK
be a low complexity of implementation and an efficient
Future 4G network is an adaptive network. More means of negotiation between the end users and the
advances in networks are needed to keep pace with the wireless infrastructure.
rapidly changing terminals and applications, as
follows[16]: REFERENCES
1. Smart antenna, software radio, together with [1] CWTS: Certified Wireless Technology Specialist Official Study
advanced base station are the key techniques to Guide. John Wiley and Sons, 29-Jun-2009 - Computers - 504
achieve adaptability of wireless access points pages.
to diverse terminals, i.e. to make radio systems [2] Wireless security and privacy: best practices and design
and air networks re-configurable. techniques By: Tara M. Swaminatha.
[3] http://www.3gafrica.org/graphical/adminews/Whitepapers/acron
2. Hierarchical and ubiquitous as well as overlay yms.pdf
cellular systems, including picocell, microcell, [4] GSM, GPRS and edge performance: evolution towards
macro-cell, and mage cell ones, implements 3G/UMTS By- Juan Melero.
seamless network interconnection of both [5] http://en.wikipedia.org/wiki/4G technology.
symmetric and asymmetric nature, and [6] A.B.M.Rezbaul Islam and Tae-Sun Chung Information and
computer Engineering Ajou University, Korea.
seamless terminal handoff of both horizontal [7] Qing Xiuhua, Cheng Chuanhui and Wang Li “ ,” Department of
and vertical levels respectively. Mathematics and Physics Wuhan University of Science and
3. Network layer hierarchical mobility Engineering. Page , year, vol.
management based on Mobile IPv6 and [8] Departement of Computer Science & Techonolgoy,Hubei
Cellular IP brings quick and seamless handoff University Of Economics.
[9] Shih-Lian Huang,”Evolution from 3G to 4G ITN-620 enterprise
to terminals. The Mobile IPv6 also presents a network design, July 2005 in ITNG.
great contribution to the adaptability of [10] J. M. Pereira. "Fourth Generation: Now it is Personal!",
heterogeneous networks. Personal, Indoor and Mobile Radio Communications, vol. 2, pp.
4. Ad hoc wireless networks are a kind of self- 1009-1016, (2000).
deployed wireless networks to make networks [11] J. Ibrahim. "4G Features", Bechtel Telecommunications
portable and adaptable, and thus dynamically Technical Journal, vol. 1, no. 1, pp. 11-14, (2002).
[12] S. Wildstrom, AT&T’s Stephenson: the road to 4G, 2009.
share unlicensed radio spectrum. [13] P. Taylor, AT&T to roll out 4G network, 2009.
5. Network reconfiguration can be obtained by [14] D. Perkins, 4G: challenges and opportunities, 2007.
the configuration of protocol stacks and [15] A. Lyle, Clear, first 4G network launched, 2009.
programmability of network nodes. Thus, it [16] http://www.cisco.com/networkers/nw00/pres/2307.pdf.
can adapt convincingly to the changing [17] http://www.agilent.com/cm/wireless/pdf/3G-
Seminar2001_02.pdf
Abstract—The goal of CMS experiment is to be injected into the LHC.This accelerator has begun to
investigate a wide range of physics, including the search generate vast quantities of data, which CERN streams
for the Higgs boson, extra dimensions, and particles that to laboratories around the world for distributed
could make up dark matter. Most of the activities at processing (making use of a specialized grid
CERN are currently directed towards operating the new
Large Hadron Collider (LHC), and the experiments for it.
infrastructure, the LHC Computing Grid). In April
2005, a trial successfully streamed 600 MB/s to seven
Keywords: Compact Muon Solenoid (CMS), Large different sites across the world. If all the data generated
Hadron Collider (LHC), Hadron Calorimeter (HCAL), by theLHC is to be analyzed, then scientists must
Cathode strip chambers (CSC), Electromagnetic Calorimeter achieve 1,800 MB/s before 2008.
(ECAL)
I. INTRODUCTION
The European Organization for Nuclear Research,
known as CERN or Cern is an international
organization whose purpose is to operate the world's
largest lephparticysics laboratory. Established in 1954,
the organization is based in the northwest suburbs of
Geneva on the Franco–Swiss border, and has 20
European member states.The term CERN is also used to
refer to the laboratory, which employs just under 2,400
full-time employees, 1,500 part-time employees, and
hosts some 10,000 visiting scientists and engineers,
representing 608 universities and research facilities and
113 nationalities. CERN's main function is to provide
the particleaccelerators and otherinfrastructure needed
for high-energy physics research-as a result, numerous
experiments have been constructed at CERN following
international collaborations. It is also the birthplace of A. Construction of the CMS detector for LHC at
the World Wide Web. The main site atMeyrin has a CERN
large computer center containing powerful data-
processing facilities, primarily for experimental data The initial particle beams were injected into the
analysis; because of the need to make these facilities LHC August 2008.The first attempt to circulate a beam
available to researchers elsewhere, it has historically through the entire LHC was at 8:28 GMT on 10
been a major wide area networking hub. September 2008,[but the system failed because of a
faulty magnet connection, and it was stopped for repairs
II. THE LARGE HADRON COLLIDER on 19 September 2008.
At 1200 BST on Tuesday 30 March 2010 the LHC
Most of the activities at CERN are currently successfully smashed two proton particle beams
directed towards operating the new Large Hadron travelling with 3.5 TeV (trillion electron volts) of
Collider (LHC), and the experiments for it. The LHC energy, resulting in a 7 TeV event. However, this is just
represents a large-scale, worldwide scientific the start of the road toward the expected discovery of
cooperation project.The LHC tunnel is located 100 the Higgs boson. This is mainly because the amount of
meters underground, in the region between the Geneva data produced is so huge it could take up to 24 months
International Airport and the nearby Jura mountains. It to completely analyse it. When the 7 TeV experimental
uses the 27 km circumference circular tunnel previously period ended, the LHC revved up to 8 TeV (4 TeV
occupied by LEP which was closed down in November acceleration in both directions) in March 2012, and will
2000. CERN's existing PS/SPS accelerator complexes begin particle collisions at that rate in early April 2012.
will be used to pre-accelerate protons which will then At the end of 2012 the LHC will be shut down for
(HCAL=hadron calorimeter, ECAL=electromagnetic optically clear material, ideal for stopping high energy
calorimeter) particles. Lead tungstate crystal is made primarily of
metal and is heavier than stainless steel, but with a
1) CMS by layers touch of oxygen in this crystalline form it is highly
transparent and “scintillates” when electrons and
photons pass through it. This means it produces light in
proportion to the particle’s energy. These high-density
crystals produce light in fast, short, well-defined photon
bursts that allow for a precise, fast and fairly compact
detector. It has a radiation length of χ0 = 0.89 cm, and
has a rapid light yield, with 80% of light yield within
one crossing time (25 ns). This is balanced however by
a relatively low light yield of 30 photons per MeV of
incident energy. The crystals used have a front size of
22 mm × 22 mm and a depth of 230 mm. They are set
in a matrix of carbon fiber to keep them optically
isolated, and backed by silicon avalanche photodiodes
for readout.The Hadron Calorimeter (HCAL) measures
the energy of hadrons, particles made of quarks and
gluons(for example protons, neutrons, pionsand kaons).
Layer 1–The Tracker Additionally it provides indirect measurement of the
presence of non-interacting, uncharged particles such as
neutrinos.The HCAL consists of layers of dense
material (brass or steel) interleaved with tiles of plastic
scintillators, read out via wavelength-
3) Layer 3–the hadronic calorimeter
The Hadron Calorimeter Barrel (in the foreground, chambers (RPC). The DTs are used for precise
on the yellow frame) waits to be inserted into the trajectory measurements in the central barrel region,
superconducting magnet (the silver cylinder in the while the CSCs are used in the end caps. The RPCs
Centre of the red magnet yoke). provide a fast signal when a muon passes through the
The high pseudorapidity region (3.0 < |n| < 5.0) is muon detector, and are installed in both the barrel and
instrumented by the Hadronic Forward (HF) detector. the end caps.
Located 11 m either side of the interaction point, this The drift tube (DT) system measures muon
uses a slightly different technology of steel absorbers positions in the barrel part of the detector. Each 4-cm-
and quartz fibres for readout, designed to allow better wide tube contains a stretched wire within a gas
separation of particles in the congested forward region. volume. When a muon or any charged particle passes
The HF is also used to measure the relative online through the volume it knocks electrons off the atoms of
luminosity system in CMS. the gas. These follow the electric field ending up at the
The brass used in the endcaps of the HCAL used to positively-charged wire. Cathode strip chambers (CSC)
be Russian artillery shells are used in the endcap disks where the magnetic field is
uneven and particle rates are high. CSCs consist of
1) Layer 4–the magnet arrays of positively-charged “anode” wires crossed with
The CMS magnet is the central device around negatively-charged copper “cathode” strips within a gas
which the experiment is built, with a 4 Tesla magnetic volume. Resistive plate chambers (RPC) are fast
field that is 100,000 times stronger than the Earth’s. gaseous detectors that provide a muon trigger system
CMS has a large solenoid magnet. This allows the parallel with those of the DTs and CSCs. RPCs consist
charge/mass ratio of particles to be determined from the of two parallel plates, a positively-charged anode and a
curved track that they follow in the magnetic field. It is negatively-charged cathode, both made of a very high
13 m long and 6 m in diameter, and its refrigerated resistivity plastic material and separated by a gas
superconducting niobium-titanium coils were originally volume. When a muon passes through the chamber,
intended to produce a 4 T magnetic field. The operating electrons are knocked out of gas atoms. These electrons
field was scaled down to 3.8 T instead of the full design in turn hit other atoms causing an avalanche of
strength in order to maximize longevity. electrons. The electrodes are transparent to the signal
The inductance of the magnet is 14 Η and the (the electrons), which are instead picked up by external
nominal current for 4 T is 19,500 A, giving a total metallic strips after a small but precise time delay. The
stored energy of 2.66 GJ, equivalent to about half-a- pattern of hit strips gives a quick measure of the muon
tonne of TNT. There are dump circuits to safely momentum, which is then used by the trigger to make
dissipate this energy should the magnet quench. The immediate decisions about whether the data are worth
circuit resistance (essentially just the cables from the keeping. RPCs combine a good spatial resolution with a
power converter to the cryostat) has a value of 0.1 mΩ time resolution of just one nanosecond (one billionth of
which leads to a circuit time constant of nearly 39 a second).
hours. This is the longest time constant of any circuit at
CERN. The operating current for 3.8 T is 18,160 A,
giving a stored energy of 2.3 GJ.
2) Layer 5–The muon detectors and return yoke
As the name “Compact Muon Solenoid” suggests,
detecting muons is one of CMS’s most important tasks.
Muons are charged particles that are just like electrons
and positrons, but are 200 times heavier. We expect
them to be produced in the decay of a number of
potential new particles; for instance, one of the clearest
"signatures" of the Higgs Boson is its decay into four
muons.
Because muons can penetrate several meters of iron
without interacting, unlike most particles they are not
stopped by any of CMS's calorimeters. Therefore,
chambers to detect muons are placed at the very edge of A part of the Magnet Yoke, with drift tubes and
the experiment where they are the only particles likely resistive-plate chambers in the barrel region.
to register a signal.
To identify muons and measure their momenta, 3) CMS
CMS uses three types of detector: drift tubes (DT), The CMS detector uses a huge solenoid magnet to
cathode strip chambers (CSC) and resistive plate bend the paths of particles from collisions in the LHC
(LHCThe Compact Muon Solenoid (CMS) is a general- its construction being started in the late 1990's. It has
purpose detector at the Large Hadron Collider).It is been finished and has already been delivered to CERN.
designed to investigate a wide range of physics,
including the search for the Higgs boson, extra VI. INDIAN CONTRIBUTION
dimensions, and particles that could make up dark
Bhabha Atomic Research Centre, which is the
matter. Although it has the same scientific goals as the
ATLAS experiment, it uses different technical solutions state-owned centre of India, announced that it was
and a different magnet-system design. developing the largest magnet of the world. This largest
magnet would weigh 50000 tons.
The magnet is said to be much bigger than the one
which is found at Compact Muon Solenoid detector at
CERN in Geneva. This magnet will play a crucial role
in the India-based Neutrino Observatory which will
come up almost 4300 feet below the cave in the
mountain in Tamil Nadu.
The Head of the atomic research centre's nuclear
physics division announced that the magnet to be
developed by the Bhabha Atomic Research Centre
would be the largest in the world in terms of its
dimensions.The magnet will be iron-based and will
weigh 50000 tons. The weight of the magnet at CERN
ranges somewhere between 4000 and 5000 tons.
V. CONCLUSION
CMS (Compact Muon Solenoid) and study about
The CMS detector is built around a huge solenoid the different Layers of CMS.CERN's main function is
magnet. This takes the form of a cylindrical coil of to provide the particle accelerators and other
superconducting cable that generates a field of 4 tesla, infrastructure needed for high-energy physics research-
about 100,000 times the magnetic field of the Earth. as a result, numerous experiments have been
The field is confined by a steel “yoke” that forms the constructed at CERN following international
bulk of the detector’s 12,500-tonne weight. An unusual collaborations. It is also the birthplace of the
feature of the CMS detector is that instead of being built World Wide Web.
in-situ like the other giant detectors of the LHC CMS is designed as a general-purpose detector,
experiments, it was constructed in 15 sections at ground capable of studying many aspects of proton collisions at
level before being lowered into an underground cavern 14 TeV, the center-of-mass energy of the LHC particle
near Cessy in France and reassembled. The complete accelerator.
detector is 21 meters long, 15 meters wide and 15
metershigh. The CMS experiment is one of the largest REFERENCES
international scientific collaborations in history,
[1] “CERN.ch". Public.web.cern.ch. Retrieved 20 November 2010.
involving 4300 particle physicists, engineers, [2] "The Name CERN". CERN. 30 September 2011. Retrieved 16
technicians, students and support staff from 179 August 2012.
universities and institutes in 41 countries [3] "CERN.ch". Public.web.cern.ch. Retrieved 20 November 2010.
A quick search indicates just how eager people are [4] "CERN.ch La". Public.web.cern.ch. Retrieved 20 November
2010.
to declare that their magnet is the world's largest [5] "CERN.ch". Public.web.cern.ch. Retrieved 20 November 2010.
without actually checking to see if anyone else has a [6] "Antihydrogen isolation". CNN. 18 November 2010.
bigger one. Sometimes if you read the fine print, you [7] Jonathan Amos [6 June 2011]BBC © 2011 Retrieved 2011-06-
06.
see things like "the world's largest magnet used for MRI
[8] "CERN.ch". Public.web.cern.ch. Retrieved 20 November 2010.
research" which is really just a subset of all the magnets [9] "W3.org". W3.org. Retrieved 20 November 2010.
in the world.Reading our question literally, I'd say the [10] "CERN.ch". CERN.ch. Retrieved 20 November 2010.
world itself is the world's biggest magnet. The field at [11] Adrian Cho, Neutrinos Travel Faster Than Light, According to
One Experiment, Science NOW, 22 September 2011.
the surface of the earth is tiny, a bit less than one Gauss,
[12] The Associated Press, "Einstein Proved Right in Retest of
but the Earth is quite big. Even just the magnetic part, Neutrinos' Speed", the Associated Press, 17 March 2012.
the metallic core, is by far the biggest magnet around. [13] "CERN Press Release". Press.web.cern.ch. Retrieved 4 July
The biggest (at least in size) magnet I know of is the 2012.
[14] "CERN Website–LINAC". Linac2.home.cern.ch. Retrieved 20
superconducting solenoidal magnet of the CMS
November 2010.
experiment on the Large Hadron Collider at CERN in [15] CERN Courier, "MoEDAL becomes the LHC's magnificent
Geneva, Switzerland. Here's a short press release about seventh", 5 May 2010
[16] Overbye, Dennis (29 July 2008). "Let the Proton Smashing [29] State”. CERN press release. CERN. Retrieved 5 October 2012.
Begin. (The Rap Is Already Written.)". The New York Times. [30] "24 June 2011: CERN-CERN Council looks forward to summer
[17] "CERN press release, 7 August 2008". Press.web.cern.ch. 7 conferences and new members". Interactions.org. 24 June 2011.
August 2008. Retrieved 20 November 2010. Retrieved 7 December 2011.
[18] "'God particle': New particle found, could be the Higgs boson, [31] "ISAAR relationship data at CERN library". Retrieved 14
CERN scientists say". 4July 2012. December 2009.
[19] ESA Convention (6th ed.). European Space Agency. September [32] "CERN International Relations-Jordan". International-
2005. relations.web.cern.ch. Retrieved 4 July 2012.
[20] "Convention for The Establishment of A European Organization [33] "CERN International Relations-SESAME". International-
For Nuclear Research". CERN Council website. CERN. relations.web.cern.ch. 17 October 2011. Retrieved 4 July 2012.
Retrieved 16 July 2012. [34] ''Macedonia joins CERN (SUP)''". Mia.com.mk. Retrieved 20
[21] "Member States". CERN Council website. CERN. Retrieved 16 November 2010.
July 2012. [35] "Shiva's Cosmic Dance at CERN". Fritjof Capra. 18 June 2004.
[22] www.newswise.com/articles/the-republic-of-cyprus-becomes-a- Retrieved 7 December 2011.
cern-associate-member-state [36] Smith, David. The Dance of Siva: Religion, Art and Poetry in
[23] "Member States Contributions-2012". CERN website. CERN. South India.Cambridge University Press. ISBN 0521528658
Retrieved 16 July 2012. 9780521528658.
[24] "Final Budget of the Organization for the fifty-eighth financial [37] "Large Hadron Collider Rap Video Is a Hit", National
year 2012". CERN. 30 November 2011. Retrieved 16 July 2012. Geographic News. 10 September 2008. Retrieved 13 August
[25] "Observers". CERN Council website. CERN. Retrieved 16 July 2010.
2012. [38] "Southparkstudios.com". South Park Studios. Retrieved 25 May
[26] Andresen, G. B. et al. (2010). "Trapped antihydrogen". Nature 2011.
468 (7324): 673–6. Bibcode 2010Natur.468..673A. [39] "Angels and Demons". CERN. Retrieved 31 January 2012.
doi:10.1038/nature09610.PMID 21085118. [40] Boyle, Rebecca (10.31.2012). "Large Hadron Collider
[27] "Vesti-Srbija zvanično postala član CERN-a". B92. Retrieved 4 Unleashes Rampaging Zombies". Retrieved 22 November 2012
July 2012.
[28] "CERN Press Release". Public.web.cern.ch. 16 September
2011. Retrieved 7 December 2011.
Abstract—In this paper the Graphite epoxy epichlorohydrin and consists of glycidol and hydroxyl
composites which is formed by very large number of groups with ether bond on the main chain. It is widely
stacked graphene monolayer & through ultrasonic used in adhesive, electronic and coating industries due
assisted dispersion of graphite nanoparticles (500 nm) at
various concentrations ranging 1.5 to 4.5wt% into
to the small volume shrinkage in curing and outstanding
commercially available epoxy resin (DGEBA(CY-230)), electrical performance. In the aerospace industry,
under heating at 900C and curing with commercially bisphenol A is commonly used as a primer matrix on
available amine hardener HY-951 (10wt%) at 400C. The aluminium alloy 2024-T3 for protecting the surface of
fabricated nano composites were characterized by UV- aircraft [3]. The incorporation of various reinforced
visible spectroscopy in N-methyl pyrolidone nanosized phases into the epoxy, such as graphite
suspension/dispersion, scanning electron microscopy nanofiber, carbon nanotubes, nanoclays, cellulose nano
(SEM) and four probes electrical conductivity at room
fiber and nano alumina resin, is one of the most
temperature. A UV-vis spectrum reveals a common
absorption at 277 nm that gets intensified with increase in effective ways to improve the properties of the epoxy
concentration of the graphite nanoparticles, indicating the [9–11]. The efficiency of nano additives for UV
physical attachment of graphite with epoxy matrix. SEM protection depends upon a number of parameters. Size
data reveals the formation of a uniformly dispersed of the particles is one of the most important factors. On
graphite nanoparticles into epoxy at 1.5wt% and onwards. incorporation of inorganic oxides into the epoxy matrix,
Graphene has shown proven result as an emblematic the UV absorption of nano sized particles was reported
protagonist, seeming to answer many need, characterized
to be much stronger than that of micro sized ones. In
by superior electrical & mechanical performance.
addition, nanoparticles with smaller size lead to higher
Keywords: Graphite; epoxy resin; hardener HY-951; N- transparency and better UV protection. A reduction of
Methyl-2-pyrrolidone; Graphite-epoxy composite electrode the refractive index is also beneficial to UV protection.
While concentration and film thickness also affect the
I. INTRODUCTION transparency, the UV protection qualities of a number
Graphite/epoxy composites are increasingly being of nano sized additive absorbers were compared [12]. It
used for the fabrication of aircraft and spacecraft & is expected that the Graphite nanoparticles would
microwave applications because of their low density enhance the UV absorption properties of the resulting
and their exceptionally high strength and modulus. nano composite thin films.
They are replacing metals, such as aluminum (Al)
alloys, which have poorer mechanical properties and II. EXPERIMENTAL WORK
higher densities. However, the replacement of metals in
The synthesis of Graphite/Epoxy nano composite
many electrical applications has proven to be difficult
because the resistivity of graphite/epoxy composites is was executed as follows:
typically three orders of magnitude higher than that of For the synthesis of Graphite/Epoxy composites,
the metals they replace [1]. Up to now, studies on the graphite powder (1.5 ~ 4.5wt %) was sonicated with
graphite/epoxy-based composites have been focused the diglycidyl ether of Bisphenol A (Epoxy) for 30
mainly on some polymer matrices such as poly-methyl minutes. Then the contents were heated at 900C
methacrylate (PMMA) [2–4] and polystyrene [5–7]. followed by cooling at 400C and then cured with tri-
Graphite fiber/polymer resin composites can ethylene tetra mine (HY-951) TETA in the form of
provide the required mechanical and physical properties panels. Then the composites were post cured for 5 hours
[8]. Epoxy resin has developed rapidly since it was at 60-700C in oven. Similarly Epoxy was cured with
invented, and has been widely used in practical TETA under identical reaction condition in absence of
applications. The most important and industrialized graphite powder and treated as a control.
epoxy is bisphenol A, which is derived from
III. UV SPECTRA OF EPOXY/ GRAPHITE COMPOSITES natural graphite in epoxy resin was conducted by high
speed mixer through ultra-sonic bath sonicator with
The study of ultra-violet spectra of the
frequency above 18 kHz (18,000 cycles per second).
graphite/epoxy composites is described below. In this
The graphite epoxy adjournment was poured into the
experiment the graphite/epoxy composite powders at
mold and compression molding was conducted for
the different concentrations (1.5 ~ 4.5wt%) were
fabrication of graphite epoxy composites. The
sonicated with the N-Methyl-2-pyrrolidone in an
conductivities of graphite/epoxy composites at different
ultrasonic probe which is used for speed dissolution, by
concentrations are shown in figure 2. Figure 3 shows
breaking intermolecular interactions between the
the I-V characteristics of the composites. Electrical
composite powder & N-Methyl-2-pyrrolidone for 60
conductivity was measured by four point probe method.
minutes for achieving proper mixing and then the
Results reveal that non-conductive epoxy polymer
contents were prepared for ultra-violet spectra analysis.
becomes conductive on addition of graphite. The
UV spectra of graphite/epoxy composites are observed
electrical conductivity increases with increase in the
at the room temperature (250C) with UV/Visible
percentage of graphite in the epoxy resin. But, in this
Scanning Spectrophotometer which included single-
experiment it has been observed that graphite/epoxy
and six-cell cuvette holders which can accommodate
composite of 1.5wt% has threshold point at which there
10-mm square cuvettes. Firstly the solution (4µL) of
are sufficient electrons to make a low resistance
graphite/epoxy composite powder & N-Methyl-2-
conducting path. It means that the electrical
pyrrolidine was put into a single-position cell in the six-
conductivity increased for composition of composition
position cell holder of the UV/visible scanning
up to 1.5wt% after that it decreased and we discuss the
Spectrometer through the Micropipette. The ratio of
temperature dependence properties of two compositions
solution between the N-Methyl-2-pyrrolidine &
in the Sr-Fe-Co-O systems, named as SFC.
graphite/epoxy composite powder was 3:1 with
SFC ≈9.99349×3E-5 S/m of epoxy
different concentration of graphite (1.5 ~ 4.5wt %) in
SFC ≈9.99365×3E-5 S/m of 1.5wt%
the epoxy resin. The UV absorption peaks and the
SFC ≈9.99361×3E-5 S/m of 3.0wt%
absorption edges of the studied composites were
SFC ≈ 9.99356×3E-5 S/m of 4.5wt%
observed in the UV-Vis absorption spectra at room
temperature. In the following Figure: 1, a strong
absorbance peak located around 277 nm. In UV Spectra
analysis, the 4.5wt% graphite/epoxy composite has the
maximum absorbance edge & this shows that the
highest absorbance intensity peak is observed at highest
concentration of graphite.
TABLE 1
Polymer/ Wavelength(λ) Absorbance
Composites in nm Units
Epoxy 277 0.131 Fig. 2: Measurement of Electrical Conductivity
1.5wt% 277 0.141
3.0wt% 277 0.171
4.5wt% 277 0.252
V. SCANNING ELECTRON MICROSCOPY OF GRAPHITE/ nanoelectronics, nanooptics and solid state physics the
EPOXY COMPOSITES AT DIFFERENT PERCENTAGES various epoxy/graphite particulate composites were
prepared with different particulate fractions and were
The distribution of graphite particles in the epoxy
characterized for the electrical conductivity, UV
resin were examined using a scanning electron
spectra, and morphological properties. In electrical
microscope (SEM). In the following SEM testing, the
conductivity experiment, it has been observed that the
fracture surfaces were examined using the JSM-
graphite/epoxy composite of 1.5wt% has the maximum
6610LV which is a scanning electron microscope
conductivity at which there is sufficient electrons to
(SEM) to identify image and analyze a wide range of
make a low resistance conducting path. It means that
conducting and non-conducting (Epoxy) samples. The
highest conductivity is measured at 1.5wt%. In U V
fracture surfaces were gold coated prior to SEM
Spectra analysis, the 4.5wt% graphite/epoxy composite
investigation to avoid charging and were examined at
showed the maximum absorbance intensity. So it
10 kV accelerating voltage. Since graphite is
demonstrates that, as the concentration of graphite
conductive, the particles could also be examined at 10
increases, the UV absorbance increases. SEM data
kV accelerating voltage without gold coating. The
reveals the formation of a uniformly dispersed graphite
following images were observed at different
nanoparticles into epoxy. The SEM images reveal a
magnification (1000x, 2700x, 5500x) & different
proper mixing of graphite at 1.5wt%, 3.0wt% & 4.5wt%
resolution (2µm, 5µm, 10µm). In the following figure 4,
into epoxy resin.
the SEM images reveal the formation of a uniformly
dispersed graphite nanoparticles at the different
ACKNOWLEDGMENT
concentration (1.5wt%, 3.0wt% & 4.5wt %) into epoxy
resin. Authors would like to thank for support from
providing facility (Dr. Jamil Akhtar Scientist G, Mr.
William Ringal Taube N. & Dr. Sanjeev kumar Gupta,)
CSIR-Central Electronics Engineering Research
Institute Pilani India
REFERENCES
[1] J.R. Gaier, P.D. Hambourger, and M.E. Slabe, Carbon 29, 313
(1991).
[2] W.G. Zheng, S.C. Wong and H.J. Sue, Polymer 43, 6767
(2002).
[3] G.H. Chen, W.G. Weng, D.J. Wu and C.L. Wu, Eur. Polym. J.
39, 2329 (2003).
[4] W. Zheng and S.C. Wong, Compos. Sci. Technol. 63, 225
(2003).
[5] W. Zheng, X.H. Lu and S.C. Wong, J. Appl. Polym. Sci. 91,
2781 (2004).
[6] J.F. Zou, Z.Z. Yu, Y.X. Pan, X.P. Fang and Y.C. Ou, Scr.
Mater. 40, 954 (2002).
[7] W.P. Wang and C.Y. Pan, Polymer 45, 3987 (2004).
[8] Evans, D.; Morgan, J. T.; Sheldon, R.; and Stapleton, G.B.: Post
Irradia-tion Mechanical Properties of Epoxy Resin/Glass
Composites. MELD-200, Sci. Res. Counc. (British), June 1970.
[9] Shi H, Liu F, Yang L and Han E 2008 Prog. Org. Coat. 62 359
CrossRef.
Fig. 4: SEM Micrographs Revealing Fracture Surfaces [10] Omrani A, Simon L.C. and Rostami A A 2009 Mater. Chem.
with 1.5wt%, 3.0wt% and 4.5wt% Phys. 114 145 CrossRef
[11] Kippax P, Waid-Smith S and Pugh D 2007 Paint Coat. Ind.
VI. CONCLUSION 23 44.
[12] Yin L, Wang Y, Pang G, Koltypin Y I and Gedanken A 2002 J.
Graphene nanoelectronics is emerging area of Colloid Interface Sci. 246 78 CrossRef PubMed.
research & many new applications in the area of
TRACK III
SIGNAL PROCESSING AND PATTERN RECOGNITION
ISBN: 978-93-82880-20-2 1
Proceedings of National Conference on Trends in Signal Processing & Communication (TSPC’13), 12th–14th April 2013
ISBN: 978-93-82880-20-2 2
Proceedings of National Conference on Trends in Signal Processing & Communication (TSPC’13), 12th–14th April 2013
Abstract—Eddy current flowmeters are developed for using CMOS and FPGA implementation [6,7].
measuring the sodium flow rate in the outlet of the According to a survey conducted on DDS for
primary sodium pump used in sodium cooled fast breeder generating sine wave, each method requires specific
reactor. A fast response system is developed for enhancing memory and significant additional circuitry [1]. In this
the measurement of Eddy current flowmeter in
monitoring the flowrate. To characterize the performance
paper a method for producing sine wave and arbitrary
of the developed system various types of test signals are waves using PSoC microcontroller for a specific
generated which will resemble the actual signal from Eddy application is discussed and it eliminates additional
current flowmeter. This paper deals with the generation of circuitry.
the various types of such signals by direct digital synthesis
method using Programmable system on chip II. DESCRIPTION OF EDDY CURRENT FLOWMETER
microcontroller. In addition the performance of the
developed fast response system was tested using the The Eddycurrent flowmeter (ECFM) consists of
generated signals. one primary and two secondary coils placed
symmetrically one in upstream side and other in
Keywords: Direct digital Synthesis; Eddy current;
PSoC; Signal generation; sinusoidal signal
downstream side of primary. The primary coil is excited
by a constant current of 200mA with a frequency of
I. INTRODUCTION 400Hz [8,9]. All the three coils are wound over the
bobbin made of magnetic material and are encapsulated
Signal generators generally have an electronic in stainless steel pocket to maintain physical separation
oscillator which are capable of creating repetitive of flowing fluid with primary and secondary coils.
waveforms such as ECFM works on the principle of change in the magnetic
sinusoidal,triangular,sawtooth,square etc.,. Modern field due to induced eddy currents as a result of sodium
devices use digital signal processors to synthesize such flow [8-13].When ECFM sensor is placed in a pipe
waveforms followed by digital to analog converter to filled with static sodium and if primary coil is excited,
produce the analog output. Arbitrary waveform then equal transformer voltage is generated in both the
generators (AWG) are another type of signal generators secondary because of equal flux linkage due to
which produce arbitrary waveforms. AWGs are more symmetry. If these two secondary coils are connected in
expensive than simple signal generators and also have phase opposition, the resultant output voltage is zero. If
limited bandwidth. the primary excited ECFM is placed in flowing sodium,
Direct Digital Synthesizer (DDS) is a type of then voltage is induced in both the secondary due to
frequency synthesizer used for creating arbitrary motion of the flowing sodium in addition to the
waveforms by generating a time-varying signal in transformer voltage [12-14]. The outputs of the ECFM
digital form by the use of phase accumulator, storing it secondary signals are sinusoidal in nature at a frequency
in a ROM and then performing a digital-to-analog of excitation (400Hz). Figure 1 shows the schematic of
conversion. Following this the signal is passed through ECFM sensor.
a filter for smoothing the generated signal [1-3]. DDS S2 = Etrans + Emotion (1)
uses Taylor series method or lookup table for S1 = Etrans - Emotion (2)
generation of signals [4,5].DDS are superior over Etrans represents the component of voltage due to
analog signal generators for their compactness and transformer action and Emotion represents the component
lower power consumption. DDS are manufactured of voltage due to motion induced voltage [13,15].
Equation (1) and (2) represents the secondary voltages is made to generate 2V signal. Figure 3 depicts the
of ECFM sensor. circuit diagram of the signal generator including PSoC
processor. Since the original ECFM signal is in
III. DESCRIPTION OF FAST RESPONSE SYSTEM millivolts range,,the generated 2V signal is fed to a
potential divider arrangement which converts this 2V
The fast response electronics has a signal
signal in to millivolts range. The output of the potential
conditioning unit, Programmable System on Chip dividers provides voltage ranges between (0 to 198mV)
(PSoC) microcontroller unit and display unit in addition and (0 to 360mV). These two signals that immitate the
to a graphical monitoring system [16,17]. Figure 2
secondary voltages of ECFM sensors, is given to each
shows the Block diagram representation of the channel of the signal conditioning circuit for testing.
developed fast response system for ECFM. The signal
conditioning unit has Instrumentation amplifier, B. Microcontroller Design
Isolation amplifier, bandpass filter and True RMS to
DC converter. Microcontroller unit has PSoC 3 Figure 4 shows the hardware design of PSoC for
processor whose hardware design includes 16 bit delta generating the test signal. The PSoC microcontroller
sigma ADC, Multiplexer, Timer,key module,LCD has programmable analog and digital blocks along with
module and UART.In in addition the microcontroller programmable interconnect which enable the user to do
unit has decoder for the key interface. application specific design. This work uses 8 bit Digital
to Analog Converter (DAC) and 16-bit timer for
IV. DESIGN REQUIREMENT generating the test signal. The sinusoidal signal is
generated with an incremental angle of 15°.Since the
The nature of ECFM signal is sinusoidal at a DAC maximum count is 255, the generation is made by
frequency of 400 Hz. The performance of the fast taking 128 as mid value. The corresponding calculation
response system at each stage of the signal conditioning for signal generation and lookup table is given in table
circuit can be determined by giving a sinusoidal signal I. The phase angle values are rounded to integer and an
at 400Hz.The actual signal from the ECFM sensor array of 24 values is created representing the phase
secondary coils are in the range of (0 to 100mV). So a value of the signal [18], as shown in equation (3). The
variable voltage sinusoidal signal needs to be generated signal should have a frequency of 400Hz; hence a timer
for testing the performance. Similarly the response time of 104µs is configured to produce a frequency of 400Hz
of the system can be identified by giving a burst of with 24 steps according to equation (4) & (5). The timer
signal with varying amplitude. The following sections period is set according to the equation (6).
describes the generation of different types of signals Step value =
using PSoC Microcontroller by the method of lookup [128,161,192,218,238,251,254,251,238,218,192,161,12
table. 8,95,64,38,18,5,0,5,18,38,64,95] (3)
= 24 × 400 = 9600 (4)
= = 104µs (5)
=
× (6)
Two switches are provided for selecting the type of
test signals. Based on the selection, the generations of
signal are made as per the algorithm described in
section V.
A. Hardware Design
Fig. 2: Schematic Representation of the Fast Response System for
The hardware consists of the PSoC 1 CY8C29466 Eddycurrent Flowmeter
processor and the 5 V power supply designed for
powering the processor. Using programming the circuit
ACKNOWLEDGMENT
This work was supported by Indira Gandhi Centre
for Atomic Research, Department of Atomic Energy,
Kalpakkam, India. The Authors are thankful for the
support extended by the Director of IGCAR and all the
scientists of FRTG, IGCAR, Kalpakkam.
[6] Daniel Rairigh, Xiaowen Liu, Chao Yang and Andrew J. [13] Prashant Sharma, S.Suresh Kumar, B.K.Nashine, R. Veerasamy,
Mason, Sinusoid Signal Generator for On-chip Impedance B. Krishnakumar, P. Kalyanasundaram, G. Vaidyanathan.
Spectroscopy IEEE, 978-1-4244-3828-0/09,Pg.No: 1961-1964 Development,computer simulation and performance testing in
[7] Bindiya Kamboj and Rajesh Mehra, Efficient FPGA sodium of an eddy current flowmeter.,Annals of Nuclear Energy
Implementation of Direct Digital Frequency Synthesizer for Vol:37(3) (2010) 332-338.
Software Radios, International Journal of Computer http://dx.doi.org/10.1016/j.anucene.2009.12.009
Applications (0975 – 8887), Volume 37– No.10, January 2012 [14] Poornapushpakala.S,Gomathy.C,
[8] Veerasamy.R,Asokane.C,Narayanan.K,Dhanasekaran.R,Swami SylviaJ.I,Krishnakumar.B,Kalyanasundaram.P - Eddy Current
nathan.K and Prabhakar.R. Eddy current flowmeter for incore Flowmeter - A Review. In Proceedings of the International
flow measurement in fast reactors. 8th National seminar on Conference on Recent Advances in Space Technology Services
Physics Technology of sensors (NSTPS-8) in Feb.2001 at & Climate Changes – 2010,pp.214 -217(2010),
IGCAR,Kalpakkam. doi:10.1109/RSTSCC.2010.5712844.
[9] Veerasamy R., Sureshkumar S., Asokane C., Sivakumar N.S, [15] Dohi, A., Oda, M., Iida, S., 1976. Development of an in-core
Padmakumar G., Dash S.K., Sreedhar B.K., Chandramouli S., flow meter for the LMFBR. In: Proceedings of International
Gurumoorthi K. and Vaidyanathan G. - Eddy Current Flow Conference on Liquid Metal Technology in Energy Production,
Sensor Development And Testing For LMFBR Sodium Pumps, Pennsylvania, pp. 796–801.
15th International Conference on Nuclear Engineering Nagoya, [16] Poornapushpakala.S,Gomathy.C,SylviaJ.I,Krishnakumar.B,Kaly
Japan, April22-26,2007,ICONE15-10213. anasundaram.P - Design of Fast Response Electronics for Eddy
[10] David E.Wiegand - Summary of an Analysis of the Eddy current Current Flowmeter. In the proceedings of 7th International
flowmeter. IEEE Transaction on Nuclear Sciences Vol 15,Issue Conference on Trends in Industrial Measurements and
1 February 1968.doi:10.1109/TNS.1968.4324826 Automation TIMA - 2011, January 6-8,2011,pp.299-303.
[11] David E. Wiegand & Charles W. Michels. Performance tests on [17] Poornapushpakala.S,Gomathy.C- Design, development and
an eddy-current flowmeter’, IEEE Transactions on Nuclear performance analysis of Signal conditioning circuit of fast
Science,16(1), 1969:192-195.doi: 10.1109/TNS.1969.4325106 response electronics for Eddycurrent Flowmeter in PFBR. In
[12] Prashant Sharma, S.K.Dash, B.K.Nashine, S. Suresh Kumar, R. International Journal on Intelligent Electronic Systems
Veerasamy,B. Krishnakumar, P. Kalyanasundaram, G. (IJIES),6(1),2012,24-30.
Vaidyanathan. Performance Prediction of Eddy Current [18] Designer’s Guide to the Cypress PSoC by Robert
Flowmeter for Sodium. In Proceedings of the COMSOL Ashby,Newnes Publication,2005.
Conference 2009 Bangalore.
Abstract—AdaBoost is a well-known ensemble data. Oza [9] proposed an approach called AveBoost2
learningalgorithm that generates weak classifiers to smooth noise. Kalai and Servedio [12] present a new
sequentially and then combines them into a strong one. boosting algorithm and prove that it can attain arbitrary
Also it shows its resistance to overfitting in low noise data accuracy when classification noise is present. An
cases, a lot of experiments [2,3,4] have shown that it is
quite sensitive on noisy data. Several modifications to
algorithm, Smooth Boosting [13], is proven to tolerate a
AdaBoost have been proposed to deal with noisy data. combination of classification and feature noise. The
Bagging and Random forests have shown their resistance data distribution skewness is penalized in the learning
to noisy data which could be explained partially by their process to prevent several hardest examples from
irregularities. We study on introducing irregularities into spoiling decision boundaries in [7]. Some other works
AdaBoost and propose a method called TRandom- focus on moving the noisy data from the training set
AdaBoost which domains AdaBoost on both low and high [10, 11].
noise dataset. We conjecture that the success of the
proposed method could be explained by regarding it as a
II. RANDOM-ADABOOST
Random Forest with an ergodic sequence of weighted base
hypotheses where the weights are generated by the Bagging [6] works by making bootstrap replicate
technique used in AdaBoost. randomly from the training set and using this as a new
Keywords-AdaBoost; Random forests; noise; sub-training set. Random forests [5] are a combination
overfitting; concentrated; Artificial Intelligence; Machine of tree predictors such that each tree depends on the
Learning; Computational Intelligence values of a random vector sampled independently and
identically distributed for all trees in the forest. Bagging
I. INTRODUCTION and Random forests have shown their concentration to
noise which can be explained partially by the
The goal of ensemble learning methods is to
irregularities in the weak classifier generating
construct a strong classifier (or hypothesis) which is a
procedure. In this paper we harbinger the idea of
collection of classifiers that generated sequentially.
irregularities into AdaBoost and propose a new
Their effectiveness relies on the strength of the base
algorithm, which we call TRandom-AdaBoost.
learning algorithm and the ability to generate
Experiments show that the proposed method is superior
distribution based on the training set to simulate the real
to AdaBoost on both low and high noise dataset and is
instance space. AdaBoost [1] is a well-known ensemble
also resist to overfit. We state from an algorithm called
classifier learning algorithms. It generates weak
Random-AdaBoost, inwhich the sub-optimal classifier
classifier in each iteration by calling the weak learning
is selected instead of thebest one as in the original
algorithm with a training set weighted based on
AdaBoost. The algorithm is proposed below.
previous classification errors. It increases the weights of
• Given Training set (xm, ym), m=1,2,….,m. xi∈
examples that the previous classifier did not classify
X
correctly. Due to its weight update technique, AdaBoost
tends to assign much higher weight to noisy instances, yi∈ { -1, +1}
Set the Irregularities Level r.
which are inconsistent with the majority of examples,
and this is more serious in its later iterations [3]. Start with distribution Dl(i)= 1/m;
Several previous studies have focused on exploring the • For t= 1,….T:
realization of ensemble methods in the presence of Train weak learner using distribution D t and
noise. Opitz and Maclin [2] illustrated the overfitting get a hypothesis set {h t}
problem of AdaBoost by a simple experiment. Jiang [8] Get a hypothesis randomly from the best r
has studied the theoretical aspects of boosting on noisy percent of the set: h t: X→R
Choose ∈ R.
Update:
() ( ( )
( )=
Where Zt is a normalization factor (chosen
such that Dt+1 will be distribution).
• Output the final hypothesis:
H(x)= sign(∑ ℎ ( ))
Algorithm I. Random Adaboost
The main difference between Random-AdaBoost
and the original AdaBoost is the new harbingerd
parameter r which is used to denote irregularities level.
In each iteration a hypothesis is selected randomly from
the best r percent of the hypothesis set. Later, this
Fig. 1: Test Error of Adaboost and Random-AdaBoost, Indicated by
parameter will be discussed jointly with noise level and Solid Line and Dashed Line Respectively, on Different Iteration
the number of the iterations. Numbers. The Horizontal Axis Indicates Noise Level Which Varies
Dietterich [3] harbingerd a method Randomization from 0 to 30 and the Vertical Axis Indicates the test Error. The
to grow the tree where at each node the split is selected Randomness Level is Fixed to 30
at random from among the K best splits. This method B. TRandom-AdaBoost
was used to improve the realization of the decision tree
algorithm C4.5 on noisy data. Their idea is similar to Random forests [5], proposed by Breiman, are a
ours but they didn’t harbinger the idea of irregularities combination of tree predictors in which a random
into AdaBoost. selection of features is used to split each node. Similar
to random forests, Random-AdaBoost selects weak
III. EXPERIMENTS classifier totally in random when irregularities level is
up to 100. But to the different, Random-AdaBoost still
In this part, experiments are implemented on remains the weighted ensemble technique to generate a
dataset “ionosphere” which is a binary problem with sequence of weightsw1,w2,…,wt on the hypotheses.
351 samples and 34 input variables. The Irregularities
Level, noise level and the number of iterations are
discussed jointly. As our goal in this paper is to explore
how this simple technique works, stump is used as the
weak learning algorithm. The dataset is divided into
training set and test set randomly on each running of the
algorithm and test errors with fixed parameters are
averaged on 100 running of the algorithm. The noisy
data is generated by reversing the labels of the instances
randomly under the noise level.
A. Experiments on Noisy Data
Figure I show the test error of AdaBoost and
Random-AdaBoost on noisy data where irregularities Fig. 2: Comparison of AdaBoost and Random-AdaBoost, Indicated
level isfixed to 30. When the iteration number is small, by Solid Line and Dashed Line Respectively, on Different
AdaBoostis generally better than Random-Adaboost. Randomness Level (Indicated by r). The Horizontal Axis Indicates
Noise Level and the Vertical Axis Indicates the Test Error. The
This is easilyunderstood that in such case the strength Iteration Number is Fixed to 500
of the selected weak classifiers play a big part in
classification ability and ones selected by AdaBoost are
stronger than Random-Adaboost. As the number of
iteration increases, the test error of AdaBoost tends to
decreases in low noise cases but increases in high noise
cases. Random-Adaboost shows similar trend but the
test error decreases faster and finally domains
AdaBoost on each noise level.
Streaming guide, the computer should have the connection speed will not support the server, there is no
following specifications to adequately manipulate point of having a fast server. Thus, choosing the fastest
digital video for streaming purposes: option available in terms of bandwidth is important
• A high end PC (Intel or AMD Processor) with because if the server does not have enough bandwidth
a processor speed at least 1.2Ghz+ for the PC. to cope with the demands of the users, then students
• At least 256MB RAM with a recommended of will receive poor video quality feedback such as jerky
512 MB. images, broken sound, and even loss of connection.
• A large and fast hard disk drive with a
recommended of 80GBs+ and an RPM of III. ADVANTAGES OF USING VIDEO STREAMING
7200+. There are many advantages of using streaming
• A good quality graphics card with at least 32 video. Some of the advantages include (a) instant play,
MB of RAM. (b) distributing live events, (c) delivering long-forms of
Most streamed video comes in one of three popular media, (d) multicasting to multiple viewers, and e) the
formats: (a) Real Networks’ Real Media (b) Microsoft easy creation of streamed files. Other advantages that
Windows Media (ASF or Advanced Streaming Format), streaming video files can offer educators are creating
and (c) Apple QuickTime. These three formats are visually driven materials that are more appealing to
proprietary in that media produced by one of the learners, helping educators handle volatile or quickly
formats. All three have specific advantages and support outdated materials (e.g., CD-ROMs, tape) that can be
the Real Time Streaming Protocol (RTSP). To stream stored into a searchable database, and create
files created in these three formats, the video files must synchronized presentations by having audio accompany
be placed on a streaming media server that has the still images, graphics, or text. However, individual
server-side software to stream the video. In addition, control of pacing is the main advantage of incorporating
that server must have enough capacity and bandwidth to streamed videos into distance learning courses. With
support the number of simultaneous video streams by streamed videos, students can access the material
multiple users. Other formats for streaming video asynchronously and independent of their location.
include MPEG-2, and more recently, MPEG-4. MPEG- Students are no longer bounded by the traditional
4 is a newer standard specifically developed to address classroom or the library to view visual materials
Web and mobile delivery. MPEG-4 is backward- provided by the instructor. With streamed videos,
compatible with MPEG-2, a video compression. students can access the visual materials at home and at
A final consideration for creating streamed videos any time. Another control element is the choice over
is the server. The RealPlayer is free and available for all which material to observe on-demand. Finally, being in
major OS platforms. The RealNetworks’ Helix Servers charge of when to start, pause, skip, and review the
are available on Windows, Linux, and Unix-based visual material is another way that students can contend
operating systems and support all major file formats, with the material. In short, the primary advantage of
including RealMedia, Windows Media, QuickTime, streaming video is the ability for students to self-pace
MP3, MPEG- 4, and others. The Helix Basic Server is their learning.
free for up to five simultaneous streams. The paid
version of Helix Universal Server is a full-featured IV. LIMITATION TO CONSIDER VIDEO STREAMING
server that allows 100 simultaneous streams. Windows
Media server and player are free, but they are only However, there are drawbacks of streaming video
available for Windows OS only. Also, both the that the instructor must consider. Inadequate bandwidth
QuickTime server and player are free. The QuickTime to retrieve streamed files is one limitation. Video
player is available for most OS platforms, but the streams can be bandwidth intensive. Depending upon
QuickTime server is supported natively in MAC OS. A the Internet traffic, users who have 56K modems or
server is needed to store the media file and software that even cable/DSL connections may have difficulty in
can stream the data over the Web. The file must be terms of retrieving and playing streamed video. Internet
uploaded to the server and linked to a web site so that congestion can cause playback delays, and even living
students can watch the video. The user selects a file and on a particular road or street can influence the reception
the streaming process begins by sending small packets of streamed video connection. In addition, competing
of information over the network to the client-side technologies to control standards is another obstacle for
computer. Once enough data has been received, the video streaming. Finally, lack of training and technical
player streams the video while the player acts like a support can prevent effective video streaming use. If
buffer and collects the data at an uneven rate from the support and training are not readily available, it is
server. A generic streaming server would require the difficult to sustain streaming video in academic
following specifications to properly house and run the institutions because of limited access to technology and
videos: 1GHz+ processor, 1GB+ RAM, large hard disk knowledgeable experts who can assist in maintaining
storage capacity, and fast network connectivity. If the and developing streamed video. Equipment
requirements, technical support, and network plug-in, and the WMS HTTP Control Protocol plug-in.
infrastructures need to be firmly in place before any With the exception of the WMS HTTP Control Protocol
quality video streaming projects can occur. Therefore, it plug-in, these plug-ins are enabled by default. The
is vital that these resources are made available to take control protocol plug-in receives the incoming client
advantage of what streaming video can offer to request, determines what action is indicated by the
education. request (for example, to start or stop streaming),
translates the request into a command form, and then
A. Cause to Study Video Indexing passes the command to the server. Control protocol
Indexing video data is essential for providing plug-ins can also return notification information to
content based access. Indexing has typically been clients if there is an error condition or a change of
viewed either from a manual annotation perspective or status. Basic networking protocols such as User
from an image sequence processing perspective. The Datagram Protocol (UDP) and Transmission Control
indexing effort is directly proportional to the granularity Protocol (TCP) are used to manage more fundamental
of video access. As applications demand finer grain tasks such as network connectivity and packet error
access to video, automation of the indexing process correction. The MMS and RTSP protocols are used in
becomes essential. Given the current state of art in combination with both the UDP or TCP protocols.
computer vision, pattern recognition and image
processing reliable and efficient automation is possible V. STREAMING SERVER SYSTEM OVERVIEW
for low level video indices like cuts and image motion A streaming media system based on Windows
properties. Media Technologies typically consists of a computer
B. Problem Analysis running an encoder (such as Microsoft Windows Media
Encoder), a server running Windows Media Services,
An analysis of existing systems enabled us to and players. The encoder enables you to convert both
determine the objectives of this project in more detail. live and prerecorded audio, video, and computer screen
Our primary goal was to develop a system to enable the images to Windows Media Format. The server running
collaborative indexing, browsing, annotation of video Windows Media Services, called a Windows Media
content between multiple groups at remote locations. In server, enables you to distribute the content over a
addition the system must support, High quality video– network. Users receive the content you distribute by
MPEG-2 files, Automatic shot detection, Hierarchical using a player, such as Windows Media Player.
video segmentation, Flexibility–different domains,
communities and metadata application profiles,
International video metadata standards such as MPEG-
7,Indexing of segments, shots, frames and regions
within frames. Streaming Protocol Used to Access
Video File
A data transfer protocol is a standardized format for
transmitting data between two devices. The type of
protocol used can determine such variables as the error
checking method, the data compression method, and
end-of-file acknowledgements. If all networks were
constructed in the same manner and all networking
software and equipment behaved similarly, only one
protocol would be necessary to handle all of our data
transmission needs. In reality, the Internet is comprised
of millions of different networks running a wide array
of hardware and software combinations. As a result, the
ability to stream digital media content reliably to clients
depends on a set of several well-engineered protocols.
The protocols used to stream Windows Media-based Fig. 2.9: Streaming Media Server Overview
content are: In a typical scenario, a user clicks a link on a Web
• Real Time Streaming Protocol (RTSP) page to request content. The Web server redirects the
• Microsoft Media Server (MMS) protocol request to the Windows Media server and opens the
• Hypertext Transfer Protocol (HTTP) player on the user's computer. At this point, the Web
Windows Media Services manages the use of these server no longer plays a role in the streaming media
protocols by using control protocol plug-ins. Windows process and the Windows Media server establishes a
Media Services includes the WMS MMS Control direct connection with the player and begins streaming
Protocol plug-in, the WMS RTSP Control Protocol the content directly to the user.
•to begin
b with jjust assign one media
file(GGHM101-1-1-1.wmv) to theet playlist.
3. If you openn the playlist ffile, you woulld see that it
has the folllowing structuure
<?wsx versionn="1.0"?>
<smil>
<media src= ="C:\WMPub\\WMRoot\GH HM101-1-1-
1.wm
mv"/>
</smil>
a) The meddia element can have a number of
attributess, one of which is the t source
destinatioon src. Theere are otheer attribute
also,one of which iss some thing clipBegin.
This attrribute causess the media file to be
timeshiftted by clipBeegin secondss when the
playlist iss played backk.
The Windows
W Meddia server caan receive coontent b) The com mplete of the attribute nam mes can be
from severral different soources. Prereccorded contennt can seen in Media
M Server ttoo.
be stored locally on thhe server or retrieved froom a • Clickk on a Publishhing Point
networked file server. Live events can be capptured • On the t right hand pane, You will w see the 5
using a diggital recordingg device and processed thrrough tabs((Moniter,Sourrce etc.
an encoderr before beinng sent to thee Windows MediaM • Cickk on source aand you shoulld see your
server for broadcast.
b Wiindows Mediaa Services cann also playlist file appeaaring in the lo ocation and
rebroadcasst content streeamed from a publishing point the elements
e of thhe playlist.
on a remotte Windows Media
M server. • If yoou click on G GHM101-1-1--1.wmv the
rightt hand pan off this widow willw display
VI. TIMEE BASED VIDEEO the playback
p detaails associated with this
With the help of the Windows 2003 Streaaming media file.
server we create
c the timee based video file means thhat we • Channging or editiing any of th he attributes
play and sttop the stream
m file by the usser choice,to create
c value here wouuld cause th he original
a time baseed video we need
n a playlisst file in whicch we playlist to get moodified. For example,
e in
set the tim
me of clipbeginn and clipend the whole prrocess this playlist detaiils window, if i we were
is followinng. assiggn a 10 seccond shift to o clipbegin
(assiigned as 00:00:10.0), we willw will see
A. Offlinee Indexing of Media
M File the following
f addeed in the playlist.
<?wsx versionn="1.0" encodding="utf-8"?>
1. Foor example, suppose
s we have
h a keywoord as
<smil>
H
Hindusm, and the
t lecture beegins 05 minuttes in
<media src= ="C:\WMPub\\WMRoot\GH HM101-1-1-
too the media file with the naame Gurnank..wmv
1.wm
mv"clipBegin= =”00:00:10.0””/>
annd last for 30 minutes.
m
</smil>
• Therefore the tuple foor this woulld be
4. On the other
o client side, the user still
s make a
Gurnank-G GHM101-1-1--1-00:05:00.0
call to thhe publishing point thus thee media file
(time is in hh:mm:ss.millisecond form mat)
is playedd as a stream fi file.
Also suppose that we have annother keyworrd as
Surdas, thee lecture for which
w it beginns at 36 minuttes in
VII. MYSQL
S PHP WEEB DATABASE:
to GHM1001-1-1-1.wmv((05 mins+30 mins of Gurnnank+
CREATE DA
ATABASE
01 mins for gap).
• Therefore the tuple foor this wouldd we Getting starteed with a dataabase means fiirst figuring
Surdas-GH HM102-1-1-1--00:36:00.0 out what
w you wannt to do. Ourr project we have create
2. Create on demaand Publishinng Point and assign a datab
base of the folllowing
thhe playlist • Courseunnit
• Assign a playlist file f to it (call • CourseunnitLecture
samplePlayylist.wsx). This step is done • Courses
automaticaally by the addd publishinggpoint • Serverinffo
wizard if you
y choose thhe option to create
c • Sessions
the playlsitt(and announccement). • Videoinddex
Table structurre for table `coourseUnit`
152 Bhag
gwant Institute of
o Technology, Muzaffarnagar,
M Uttar Pradesh, India
A Video Indexing System
Abstract—This paper aims at optimization of techniques, which is based on statistical codes which
parameters which influence the compression ratio and size reduces amount of test data stored on ATE and
of decoder. To achieve that various compression and
transferred to SOC.
decompression techniques for data compression based on
statistical codes are used, which are stored on the tester Transferring compressed amount of test data will
and then applied to each core in system-on-chip. take less time in comparison of the transferring full test
In this paper we emphasized on four compression and vector at given channel capacity of tester. In case of
decompression techniques based on statistical codes, decompression a very small decoder circuitry are
which are Huffman coding, Selective Huffman coding, required on chip without loss of any actual test set. In
Optimal selective Huffman coding, Selected Selective this paper we presented compression and
Huffman coding. In case of Huffman coding, when we
increase the block size we get better compression ratio
decompression scheme that can generate deterministic
with increase in decoder size which depends on block size, scan vectors and also sent the actual sequence of scan
for different benchmark circuits For reducing decoder size vector to CUT. It does not require any prior knowledge
and increasing compression ratio selected selective of CUT, so it is suitable for testing intellectual property
Huffman coding is preferred over other techniques. cores where core supplier does not provide any
Keywords: IP core based SoC, ATE, Huffman code, information about internal structure of core.
Test data compression, Optimal selective Huffman,
I. INTRODUCTION
Test data compression involves compression of test
data both (stimulus and response) which is stored on the
ATE. Before going to scan chain some additional on-
chip hardware is added for decompression of
compressed data set which comes from ATE and output
response is compressed and applied to ATE again.
Since Tester has limited speed, channel capacity and
memory, cost of testing and test application time will
increase considerably. To solve this problem a various
compression techniques are adopted in SOC era. These Fig. 1: Block Diagram for Illustrating Scheme for a Slower
techniques are–Linear decompression based scheme, Tester Clock
broad-scan-based scheme[1],and code-based
As Fig 1 shows that data comes from tester are
scheme[2],[8]. Code based scheme is appropriate to
compressed and before going to scan chain additional
compress the pre-computed test set without requiring on chip hardware is required to decompress the data
any structural information of IP cores. Its different from then apply to the scan chain then output response
BIST and hybrid BIST because whatever test vector is obtained by scan chain is again compressed by
applied to CUT that are exactly same as before decompresser. Advantage of test data compression is
compression [5]. So with the help of test data generate the complete set of test set patterns applied to
compression memory required for ATE is reduces and the CUT and this test pattern is optimizable for desired
also test time reduces because fewer amounts of data is fault coverage.
transmitted between ATE and chip [4]. This paper Compression ratio=(1-(compression bits) (Mintest
presents various compression and decompression bits))*100 eq.1
II. METHOD FOR DON’T CARE BIT codes.At first test vector T divided into [T]/l blocks of
FILLING TECHNIQUE length ‘l' then we get k distinct
blocks,b1,b2…………..bk. with occurrence of
A large amount of don't care bits is generated by
frequency p1,p2,………=pk. In order to generate the
ATPG which gives the test data for CUT[2]. So when Huffman code, a binary tree is constructed beginning
we manipulated these don’t care bits then it increase with leaves and moving towards the root. For every
compression of test data For the statistical codes, test distinct block bi, a leaf node is generated, and a weight
data is divided into equal size blocks of B bits. To equal to pi is assigned to it (pi is the occurrence
improve the test data compression, the no. of distinct frequency of block bi). The pair of nodes with the
blocks in a given test set should be reduced and smallest weights is selected first and a parent node is
frequency of occurrence for each distinct block should generated with weight equal to the sum of the weights
be increased. In this, an algorithms to fill don't care bits of these two nodes. This step is repeated iteratively until
which has less computational complexity are present. only a single node is left unselected, the root (we note
that each node can be selected only once). Then,
A. Hamming Distance based Approach starting from the root, all nodes are visited once and the
In this approach, test data set are divided in to logic 0 (logic 1) value is assigned to each left (right)-
blocks of size B bits. These blocks may be completely child edge.
specified or partially specified i.e. with don't care bits. TABLE 1.1: TEST SET DATA PARTITIONING AND OCCURRENCE FREQ
For coding process, for each distinct block, the Test Set Distinct Block FREQ.
corresponding frequency of occurrence is calculated. At 1010 0000 1010 1111 1010 9/20
first we search the most frequently occurring block and 1111 0000 1010 0001 0000 5/20
then compare with next most frequently occurring block 1010 0000 0010 1010 1111 3/20
0000 1010 0000 1010 0001 2/20
if there is no conflict in any bit position then hamming 1010 1111 1010 0001 0010 1/20
distance between these two blocks is zero The
Hamming distance is 1 if the bits on the same position Consider at test set of 80 bit data which is given in
of two blocks are opposite, i.e. '1' and '0'. The Hamming Table 1.1 is partitioned into 4 bit fixed block so we get
distance between two blocks is summation of such bits 20 occurrences of frequency for 5 distinct blocks. When
with opposite values. The Hamming distance between we encoded that data with the help of Huffman coding
00XX (F1) and 0X1X (F2) is 0.So F1 is merged with F2 scheme then compressed bit is 40 given in Table 1.2
and result is 001X. If the Hamming distance between then compression ratio is 50% for test set but area
required for decompression is large because it depends
B1 and B2 is more than 0, the Hamming distance with
on the block size as block size increases decoder size
next block with descending order of frequency will be
become more complex.
calculated. Two blocks for which the Hamming
distance is 0, will be merged and a new block F1 will TABLE 1.2: COMPRESSED TEST SET
come into existence. The next block in the sequence Compressed Test Set
will be than compared with merged block F1. This 0100111
process is repeated until further merging is not possible. 111100
01011000
The process is repeated with the next highest frequently 100100
occurring still unmerged block. The merging has 011101101
increased the number of specified bits. Still there is a
chance that few bits are unspecified. Such bits are B. Selective Huffman Coding
replaced with zeroes So for reducing the decoder complexity problem
we implement selective Huffman coding. In this at first
III. COMPRESSION TECHNIQUES test set divided into k distinct blocks b1,b2……..bk..
In this section we will discuss about various After that distinct blocks which have most occurrence
compression techniques based on statistical codes. In of frequency is encoded b1…..bm(m<k) and remaining
statistical codes variable codeword length are used to blocks are unencoded(bm…..bk).For distinguish
represent fixed block of bits in test data. At first test set between coded and unencoded blocks each codeword is
is divided in to symbols of bits then each fixed block is preceded by a bit (1) and (0) with unencoded blocks. So
represented by variable codeword which depend on the for previous table if there is m=3 encoded blocks
occurrence of frequency of symbol. compression ratio is 28.0%), (Table1.3).
Demerit of selective Huffman coding is it required
A. Huffman Coding extra bit equally lengthens all data in compressed test
A Huffman code [4] is an optimal statistical code set (encoded or unencoded), irrespective there
that is proven to provide the shortest average codeword occurrence of frequency but its decoder circuitry is less
length among all uniquely decodable variable length complex in comparison of Huffman coding.
According to this algo. m+1 Huffman codewords With the help of various compression techniques
are used m for encoding the most frequently occurring test set are compressed and stored on ATE. After that
blocks and 1 for unencoded blocks. The occurrence compressed data applied to the decoder on chip which
frequency of the latter codeword is equal to the sum of is decompress the compressed data and apply to the
the occurrence frequencies of all the unencoded distinct CUT. For any compression scheme on chip decoder
blocks. Consider the test set of table 1.1and m=3 that is, size is directly proportional to the decoder FSM state of
0001 and0010 are unencoded blocks. The sum of the particular coding technique. So, for Huffman coding for
occurrence frequencies of 0001 and 0010 is equal to given table1.1 FSM states 2b-1 are required where b is
2/20 + 1/20 = 3/20. The proposed Selective Huffman block size.
encoding as well as the compressed test set are given in In case of selective Huffman coding and optimal
Fig 1.3. We observe that now, for encoding distinct selective huffman coding m+b states are required where
blocks 1010 and 0000, only 1 and 2 bits are respectively m is encoded block and b is block size. At last for
required, while the cost for distinct block 1111 remains selected selective Huffman coding one more channel
constant (3 bits). Only before the unencoded data from ATE indicates the status of bit stream either it is
blocks, a 3-bit codeword, instead of a single bit, is encoded or unencoded. So for that on 2 to 1 Mux is
utilized. The compression ratio in this case is equal to used in chip, one input is Din and another input is E/N
38.8 percent, which is significantly higher than that of
to Mux. If E/N is 1 then Din is encoded and it apply to
selective Huffman coding.
decoder and if E/N is ‘0’ then Din is unencoded and
apply to serializer
Note-: Here it is clarify that data compression
depends upon the frequency distribution of the symbol
and no of encoded symbol selected. And if no of
encoded block is change then FSM will also change. So
conclusion is that:
1. Decoder size is directly proportional to the
decoder FSM states in any type of Huffman
coding.
2. So for FSM states decoder depends on the
frequency distribution of symbol of given test
set.
Fig. 3: Optimal Selective Huffman Coding
TABLE 1.3: EXAMPLE FOR HUFFMAN, SELECTIVE, OPTIMAL, SELECTED SELECTIVE HUFFMAN CODE
Distinct Pattern Freq. Huffman Code Selective Optimal Selective Huffman Selected Selective
Huffman Code Code Huffman Code n-4
n=3 n=3 n=3
blocks 4 4 4 4 4 E/N
S0-1010 9/20 0 10 0 0 ‘1’
S1-0000 5/20 10 110 10 10 ‘1’
S2-1111 3/20 111 111 110 11 ‘1’
0001 2/20 1101 00001 1110001 0001 0
0010 1/20 100 00010 1110010 0010 0
Test set 80 40 57 49 37
Compression (%) 50% 28.7% 38.75% 53.75%
TABLE 1.4: COMPARISON OF TEST DATA COMPRESSION RATIO
Circuit Mintest Huffman Selective Optimal Selected
N=8 encoded block N=8 encoded block N=8 encoded
block
block size=4 block size=4 block size=4 block size=4
S5378 23754 48.257 29.14 46.37 54.14
S9234 39273 52.385 30.1 51.4 55.5
TABLE 1.5: COMPARISON OF SYNTHESIS RESULT FOR DECODER OF HUFFMAN, SELECTIVE, OPTIMAL, AND SELECTED SELECTIVE HUFFMAN CODE
ISCAS5378 Huffman Selective Optimal Selected
States 15 12 12 7
Transitions 26 18 18 17
Register utilization 8 12 14 7
No of slices 8 out of 960(0%) 16 out of 960(1%) 16 out 960(1%) 6 out of 960(0%)
Slice F/F 8 out of 1920(0%) 12 out of 1920 13 out 1920(0%) 7 out of 1920(0%)
( 0%)
4 input LUTS 17 out of 1920(0%) 31 out of 1920 (1%) 31 out 1920(1%) 12 out of1920(0%)
I/Os 7 out 66(10%) 7 out of 66 (10%) 7 out 66(10%) 8 out of 66(12%)
Clock frequency 319.183MHz 217.486MHz 216.82 MHz 342.58MHz
As we are implemented all four techniques for [4] F.G. Wolff, C. Papachristou, “Multiscan-based test compression
and hardware decompression using LZ77,” in Proc. Int’l Test
decoder in VHDL and simulate at modelsim, Synthesis Conf., Oct. 2002, pp. 331-339.
shows that decoder size of selected selective Huffman [5] V. Iyengar, K. Chakrabarty, and B.T. Murray, ”Huffman
coding technique is less complicated than other and this encoding of test sets for sequential circuits,” IEEE Trans
technique requires minimum FSM states for same Instrumentation and Measurement, vol. 47, no. 1, pp. 21-25,
Feb. 1998.
encoded blocks for all techniques [6] X. Kavousianos, E. Kalligeros, D. Nikolos, “Test Data
Compression Based on Variable-to-Variable Huffman Encoding
REFERENCES With Codeword Reusability,” IEEE Trans Comput.-Aided Des.
Integr. Circuits Syst., vol. 27, no. 7, pp. 1333-1338, Jul. 2008.
[1] K. J. Lee, J. J. Chen, C. H. Huang, ”Broadcasting test patterns to [7] L. Lingappan, S. Ravi, A. Raghunathan, N.K. Jha, and S.T.
multiple circuits,”IEEE Trans. Computer-Aided Des.Integra Chakradhar, “Test-Volume Reduction in Systems-on-a-Chip
Circuits and Syst., vol. 18, no. 12, pp. 1793-1802, Dec. 1999. Using Heterogeneous and Multilevel Compression Techniques,”
[2] H. Ichihara, Y. Setohara, Y. Nakashima, and T. Inoue, “Test IEEE Trans. Comput.-Aided Des. Integr. Circuits Syst., vol. 25,
Compression/decompression based on JPEG VLC algorithm, in no. 10, pp. 2193–2206, Oct. 2006.
Proc. Asian Test Symposium., Oct. 2007, pp. 87-90. [8] M.H. Tehranipour, M. Nourani, K. Arabi, and A. Afzali-Kusha,
[3] A. Wurtenberger, C.S. Tautermann, and S. Hellebrand, Data ”Mixed RL-Huffman encoding for power reduction and data
Compression for multiple scan chains using dictionaries with compression in scan test,” in Proc. IEEE Symp. Circuits and
Corrections,” in Proc. Int’l Test Conf. (ITC 2004), pp. 926-935, System. vol. 2, pp. 681-684, May 2004.
Oct. 2004.
TRACK IV
NETWORK AND COMMUNICATION
ISBN: 978-93-82880-20-2 3
Proceedings of National Conference on Trends in Signal Processing & Communication (TSPC’13), 12th–14th April 2013
Prediction of Mobile
User behavior using Clustering
Sandhya Avasthi1 and Avinash Dwivedi2
1
Computer Science Department, Krishna Engineering College, Ghaziabad, India
2
Computer Science Department, Krishna Engineering College, Ghaziabad, India
e-mail: 1sandhya_avasthi@yahoo.com, 2avinash_dwivedi@rediffmail.com
Abstract—The services which are provided to the of meters and the number of base stations are usually
wireless mobile devices (such as PDAs, Cellular Phones, more than 10,000 in a city. When users move within the
and Laptops) from anywhere, at any time using ISAP mobile network, their locations and service requests are
(Information Service and Application Provider) are stored in a centralized mobile transaction database [2],
increased by mining and prediction of mobile user
behaviors. But these discovery may not be always be good
[3]. Obviously, the behavior pattern, in which the
enough for predictions since the differentiated mobile location and the service are inherently coexistent, of
behaviors among users and temporal periods are not mobile users becomes more complex than that of the
considered simultaneously in the previous works. User traditional web systems [4].
relations and temporal property are used simultaneously Fig. 1 shows an MC scenario, where a user moves
in this work. Here CTMSP-Mine (Cluster-based Temporal in the mobile network and requests services in the
Mobile Sequential Pattern-Mine) algorithm is used to corresponding cell through the mobile devices. Fig. 1a
mine CTMSPs. In CTMSP-Mine requires user clusters, shows a moving sequence [2] of a user, where cells are
which are constructed by Cluster-Object-based Smart
Cluster Affinity Search Technique (CO-Smart-CAST) and
underlined if services are requested there. Fig. 1b shows
similarities between users are evaluated by Location- the record of service transactions, where the service s1
Based Service Alignment (LBS-Alignment) to construct was requested when this user moved to the location A at
the user groups. The temporal property is used by time time 5. In fact, there exists insightful information in
segmenting the logs using time intervals. The specific time these data, such as movement and transaction behaviors
intervals to segment the huge data logs are found using of mobile users. Mining mobile transaction data can
Genetic Algorithm based method called GetNTSP (Get provide insights for various applications, such as
Number of Time Segmenting Points). The user cluster
information resulting from CO-Smart-CAST and the time
prediction of subsequent locations visited by user and
segmentation table are provided as input to CTMSP-Mine user’s service requests and service recommendations
technique, which creates CTMSPs. The prediction [5].
strategy uses these patterns to predict the mobile user next A mobile transaction database is complicated since
behavior. a huge amount of mobile transaction logs is produced
Keywords: Mining methods and algorithms, mobile based on the user’s mobile behaviors. Data mining is a
environments, Location Based Services, mobile sequential widely used technique for discovering valuable
patterns information in a complex data set and a number of
studies have discussed the issue of mobile behavior
I. INTRODUCTION mining. To achieve a quick response from the system,
data mining, which has been used successfully in many
Due to the popularity of mobile devices, mobile
applications, is one of the most promising technologies
users can requests services through their mobile devices
used to fulfill a dynamic service request. Previous
via Information Service and Application Provider
works addressed the problem of mining associated
(ISAP) from anywhere anytime [1].The advancement of
wireless communication techniques and the popularity service patterns in mobile web network [4]. Previous
of mobile devices such as mobile phones, PDA, and works also proposed methods to efficiently mine users’
GPS-enabled cellular phones, have contributed to a new sequential mobile access patterns, based on the FP-
business model. This business model is known as Tree.
Mobile Commerce (MC) [2] that provides Location- Path traversal patterns for mining mobile web user
Based Services (LBS) through mobile phones. MC is behaviors have been proposed. To increase the accuracy
expected to be as popular as e-commerce in the future of predictions, the moving path was taken into
and it is based on the cellular network composed of consideration in the above studies. However, mobile
several base stations. The communication coverage of behaviors vary among different user clusters or at
each base station is called a cell as a location area. The various time intervals. The prediction of mobile
average distance between two base stations is hundreds behavior will be more precise if it is possible to find the
corresponding mobile patterns in each user cluster and
time interval. This is the first work on mining and and a time interval table are generated, respectively.
prediction of mobile behaviors associated with user Third, the CTMSP-Mine algorithm is used to mine the
clusters and temporal relations [6]. CTMSPs from the mobile transaction database
according to the user cluster table and the time interval
II. ANALYSIS table which are essential in discovering the complete
information concerning personal mobile behaviors. The
A. Existing System entire procedure of CTMSP-Mine algorithm can be
In a mobile network consisting of cells with a base divided into three main steps: 1) Frequent-Transaction
station for each, users of wireless mobile devices move Mining, 2) Mobile Transaction Database
from one location to another in a random manner. The Transformation and 3) CTMSP Mining.
mobile users are served by ISPs and ISAP to access the There are three prediction strategies for selecting
World Wide Web, to get necessary information in their the appropriate CTMSP to predict the mobile behaviors
daily life. When user’s movement and their service of users: 1) the patterns are selected only from the
requests are predicted in advance, it helps to provide corresponding cluster a user belongs to; 2) the patterns
customized and efficient service to the users [3]. To are selected only from the time interval corresponding
help the user get desired information in a short time is to current time; and 3) the patterns are selected only
one of the promising applications, especially in the from the ones that match the user’s recent mobile
mobile environments, where the users do not have behaviors. If there exist more than one pattern that
much time to surf the web pages. Efficiency is satisfies the above conditions, we select the one with
increased to help mobile users experience the usage of the maximal support.
web applications and web pages as if they access from a
PC. The Existing system for prediction uses the moving
paths of users or the time a user requests for a service.
This system does not consider groups of users in
mining, but it considers only individual users. This did
not provide efficient Prediction of mobile user behavior
and it consumes more time to predict and also it lacks in
accuracy. Therefore a new system is proposed to solve
the problems in prediction.
B. Limitations of Existing System
• Most of prediction systems only consider
location as prediction parameter.
• The prediction results are inefficient.
Fig. 1: Mobile Transactions (a) Moving Sequences,
• Prediction process consumes more time. (b) Mobile Transaction Table. [6]
• No precise prediction of mobile user behavior.
D. Proposed System Benefits
C. Proposed System
• It is very precise and efficient
A novel method, named Cluster-based Temporal • It consumes less time.
Mobile Sequential Pattern Mine (CTMSP-Mine)[6], for • Considering User clusters and time
discovering CTMSPs in LBS environment is proposed. segmentation simultaneously, complete
In addition, novel prediction strategies are proposed to information concerning personal mobile
predict the subsequent user mobile behaviors using the behaviors is predicted.
discovered CTMSPs. When mobile users move within
the mobile network, the information which includes
time, locations, and service requests will be stored in
the mobile transaction database. In the data mining
mechanism, two techniques and the CTMSP-Mine
algorithm are designed to discover the knowledge. First,
the CO-Smart-CAST algorithm is proposed to cluster
the mobile transaction sequences. In this algorithm, the
LBS-Alignment is used to evaluate the similarity [7] of
mobile transaction sequences. Second, a Genetic
Algorithm based time segmentation method called
GetNTSP (Get Number of Time Segmenting Points) to
find the most suitable time intervals [8] is introduced.
After clustering and segmentation, a user cluster table Fig. 2: Overall Proposed System [6]
1) LBS alignment algorithm satisfied ones as a set of the time point sequence. In the
time point sequence, calculate the average time distance
Input data include two mobile transaction
a between two neighboring time points, calculate the
sequences [5]. Output data are the similarity between
number of neighboring time point pairs, in which the
two mobile transaction sequences, with the degrees in
time distance is higher than a. The result represents the
the range from 0 to 1. The base similarity score is set as
time segmentation count.
0.5.Use dynamic programming to calculate Mi, j where
Mi, j indicates the value of matrix M in column i and row 4) CTMSP mining
j, where M is the score matrix of LBS-Alignment. If the
locations of two transactions are the same, both the time User cluster from the second module and time
penalty and the service reward are calculated to interval table from the third module are used as input in
measure the similarity score. Otherwise the location this module. User clustering and temporal property are
penalty is generated to decrease the similarity score. considered simultaneously, such that the complete
Finally, Ms: length; s0: length is returned as the mobile sequential patterns are discovered. This module
similarity score of the two mobile transaction generates patterns called CTMSPs. The patterns and
sequences. recent behavior of users with current time are provided
to the Prediction Strategy. The prediction methods
2) The CO-Smart-CAST algorithm provide the mobile user’s behavior in the near future.
The input data are an N-by-N similarity matrix S. There are three prediction strategies for selecting
The output data are the clustering result. CO-Smart- the appropriate CTMSP to predict the mobile behaviors
CAST can automatically cluster the data according to of users: 1) the patterns are selected only from the
the similarity matrix without any user-input parameter. corresponding cluster a user belongs to; 2) the patterns
First, the CAST method [9] that takes a parameter are selected only from the time interval corresponding
named affinity threshold t is used as the basic clustering to current time; and 3) the patterns are selected only
method. Second, use a quality validation method, called from the ones that match the user’s recent mobile
Hubert’s Γ Statistics, to find the best clustering result. behaviors. The Cluster Temporal Mobile Sequential
Third, use a hierarchical concept to reduce the sparse Pattern-Mine algorithm can be divided into three main
clusters. For a clustering result, use Hubert’s Γ steps:
Statistics to measure its quality by taking the similarity • Frequent-Transaction Mining,
matrix and the clustering result as the input. In each • Mobile Transaction Database Transformation,
clustering result, calculate its Γobj and Γclu which
• CTMSP Mining.
represent the clustering qualities measured by the
The Frequent-Transaction mining finds the
original object similarity matrix S and the last cluster
similarity matrix S0, respectively. The initial values of frequent transactions that are similar. The mobile
S0 and S are the same since let every object be an transaction database is then transformed to remove the
independent cluster. noise and outliers. The CTMSP-Mine technique is used
Use the F1 score which is the harmonic mean to to discover the Cluster Temporal Mobile Sequential
combine Γobj and Γclu as ΓCO. A higher value of ΓCO Patterns. This CTMSP-Mine[6] technique uses the
represents the better clustering quality. To determine results of CO-Smart-CAST algorithm and GetNTSP
the most suitable t, the easiest way is varying t with a algorithm. The CO-Smart-CAST provides the user
fixed increment and iterating the executions of CAST to clusters [8]. Each user cluster is made as an entry in the
find the best clustering result with the highest ΓCO. User cluster table. The GetNTSP algorithm provides the
The main drawback of this way is that many time intervals that have similar mobile user behavior.
iterations of computation are required. For this reason, Each time interval is made as an entry in the Time
try to reduce the number of computations by interval table.
eliminating unnecessary executions, and then, obtain a The User cluster table and the Time interval table
“near-optimal” clustering result. That is, try to perform are provided as input to the CTMSP-Mine Technique.
a minimal number of CAST executions. The Cluster Temporal Mobile Sequential Patterns are
generated as output of the CTMSP-Mine technique.
3) The Get NTSP algorithm
The prediction strategy takes the mined patterns
The input data are a mobile transaction database D CTMSPs and the current transaction of the specific user
and its time length T. The output data are the number of (like current time, current location, current service) as
time segmenting points. For each item, accumulate the input. The results of Prediction Strategy provides the
total number of occurrences at each time point. Draw a future mobile behavior (like location, service) of the
curve of count distribution. Count occurrences of all specific user. To select the CTMSPs to provide as input
these time points, and find out the satisfied time points to the Prediction Strategy: The user clusters to which
whose counts are larger than or equal to the average of the user belongs to, is considered. The CTMSPs that are
all occurrences from these ones, and then, take these
generated from such user cluster are used as input. The intervals will have mobile users with similar behavior.
CTMSPs that are generated from the time interval For example, the users who access the information
which matches the current time are selected as the input about restaurants in the evening. This example have
to the Prediction Strategy. similar users who access the restaurant information.
Such users fall in the same time interval. The GetNTSP
IV. OVERVIEW [7]technique uses the idea behind the genetic algorithm.
The user cluster information resulting from CO-Smart-
A. Problem Definition
CAST and the time segmentation table are provided as
The problem we are addressing in the proposed input to CTMSP-Mine technique, which creates
system is formulated as follows: Given a user’s current CTMSPs. The prediction strategy uses the patterns to
mobile transaction sequence S and the current time tc, predict the mobile user behavior in the near future. The
the objective is to develop a framework to predict the mobile behavior prediction helps ISPs [1] to improve
mobile subsequent behaviors. The aim is to predict the their QoS given to mobile users.
subsequent mobile behaviors using not only S and tc but
also using all the mined CTMSPs. V. SIMULATION
The problem of CTMSPs mining is formulated as
follows: Given a mobile transaction database D A. Simulation Tool for Prediction
containing a large number of mobile transaction The Data Mining Simulation tool matlab 10is used.
sequences of users and a specified support threshold, This simulation tool could be used in multiple
the problem is to discover all the CTMSPs existing in platforms. This tool could utilize data accesses such as
the database. In this paper, the CTMSP-Mine algorithm ODBC, Excel, plain ASCII files. The data
and the behavior prediction mechanism are introduced preprocessing methods used here are pick mix,
for solving the problem of prediction. sampling, partitioning, field reordering, table fusion,
recoding numeric intervals. For clustering, K-Means
B. Overview of the Project method is used. The association rules are mined using
The services which are provided to the wireless Apriori [16] method. The visualization methods are
mobile devices (such as PDAs, Cellular Phones, and statistical graphics, 3D, animation, interactive and
Laptops) from anywhere, at any time using ISAP hierarchical navigation. To implement different
(Information Service and Application Provider) are algorithms Java API could be used.
enhanced by mining and prediction of mobile user
behaviors [13]. This business model of mobile services VI. CONCLUSION
is referred as Mobile Commerce. In this paper, a new method is proposed and named
Mining and prediction of mobile movements and as CTMSP-Mine, for discovering CTMSPs in LBS
associated transactions is the core of the project. The environments. Further, prediction strategies to predict
project focuses on discovering mobile patterns from the the subsequent user mobile behaviors using the
whole logs. But such discovery may not be precise discovered CTMSPs are introduced. In CTMSP-Mine
enough for predictions since the differentiated mobile technique, transaction clustering algorithm named CO
behaviors among users and temporal periods are not Smart-CAST is used to form user clusters, based on the
considered simultaneously in the previous works. User mobile transactions using the proposed LBS-Alignment
relations and temporal property are used simultaneously similarity measurement. Then, GetNTSP method is
in this work to provide more accuracy, and scalability. utilized to generate the most suitable time intervals.
Prediction strategy is used to predict the subsequent Using temporal periods and user clusters
mobile behavior [13]. Here CTMSP-Mine (Cluster- simultaneously for prediction, enhances the prediction
results. Such prediction results are used by the
based Temporal Mobile Sequential Pattern-Mine)
corresponding mobile service providers to enhance their
algorithm is used to mine CTMSPs. In CTMSP-Mine
services.
requires user clusters, which are constructed by Cluster-
Object-based Smart Cluster Affinity Search Technique
FUTURE WORK
(CO-Smart-CAST) and similarities [8] between users
are evaluated by Location-Based Service Alignment Implementing prioritization, make it possible to
(LBS-Alignment) to construct the user groups. provide priorities for selected users among the complex
The temporal property is used by time segmenting user behavior. Many users utilize the mobile services
the logs using time intervals [7]. The specific time every day but their interest and priorities are different
intervals to segment the huge data logs are found using from other user. Such users are prioritized over other
Genetic Algorithm based method called GetNTSP (Get mobile users. These prioritized services help to satisfy
Number of Time Segmenting Points). The time the needs of mobile users completely when resources
are limited.
I. INTTRODUCTION
Bluetoooth is a technology for shhort range wirreless
data and real time two-way auddio/video traansfer
providing data rates up to 24 Mbps.. It operates at a 2.4
GHz frequuency in the free Industriial, Scientific, and
Medical (ISM) bannd. Bluetootth devices that
communicaate with each other form a Pico P net.
The deevice that inittiates a connecction is the piconet
master andd all other devices
d withinn that piconeet are
slaves. Thee radio frequency (RF) waaves can peneetrate
obstacles, because of thist reason thhe use of wirreless Fiig. 1: No Mobile Carrier Fees Reqquired, no Interneet Required
communicaation systemss have grownn rapidly in recent r
A scatternet is formed w when the dev vices act as
years.
'master' or 'slave'' devices in m multiple picoonets at the
The wireless
w devicces can comm municate witth no
samee time. The PINs
P often coontain only foour decimal
direct linee-ofsight betw ween them. This makess RF
digitts, the strengtth of the resuulting keys is not enough
communicaation easier to use than wiredor inffrared
for protection against passsive eavesdro opping on
communicaation, but it also
a makes avvesdropping easier.
e
Commmunication.
Moreover, it is easier to t disrupt andd jam wirelesss RF
ISBN: 978-9
93-82880-20-2 167
Proceedings of National Conference on Trends in Signal Processing & Communication (TSPC’13) 12th–14th April 2013
Step 3: The claimant returns the most significant address (BD_ADDR), the 128-bit random number
32 bits of the E1 output as the computed (EN_RAND), a slot number based on the Pico net
response, the Signed Response (SRES), to clock, and an encryption key, which when combined
the verifier. initialize the LFSRs before the transmission of each
Step 4: The verifier compares the SRES from the packet, if encryption is enabled.
claimant with the value that it computed.
Step 5: If the two 32-bit values are equal, the
authentication is considered successful. If
the two 32-bit values are not equal, the
authentication fails.
Performing these steps once accomplishes one-way
authentication. The Bluetooth standard allows both one-
way and mutual authentication to be performed. For
mutual authentication, the above process is repeated
with the verifier and claimant switching roles.
If authentication fails, a Bluetooth device waits an
interval of time before making a new attempt. This time
interval increases exponentially to prevent an adversary
from attempting to gain access by defeating the
authentication scheme through trial-and-error with
different link keys.
B. Confidentiality Fig. 4: Bluetooth Encryption Process
Protecting information from eavesdropping by The encryption key (KC) is derived from the
ensuring that only authorized devices can access and current link key and may vary in length in single byte
view transmitted data. In addition to the Security Modes increments from 1 byte to 16 bytes in length, as set
for pairing and authentication, Bluetooth provides a during a negotiation process that occurs between the
separate confidentiality service to thwart attempts to master and slave devices. During this negotiation, a
eavesdrop on the payloads of the packets exchanged master device makes a key size suggestion for the slave.
between Bluetooth devices. Bluetooth has three The initial key size suggested by the master is
Encryption Modes, but only two of them actually programmed into the controller by the manufacturer and
provide confidentiality. The modes are as follows: is not always 16 bytes. In product implementations, a
Encryption Mode 1: No encryption is performed on “minimum acceptable” key size parameter can be set to
any traffic. prevent a malicious user from driving the key size down
Encryption Mode 2: Individually addressed traffic to the minimum of 1 byte, which would make the link
is encrypted using encryption keys based on individual less secure.
link keys; broadcast traffic is not encrypted.
Encryption Mode 3: All traffic is encrypted using C. Authorization
an encryption key based on the master link key. Allowing the control of resources by ensuring that
Encryption Modes 2 and 3 use the same encryption a device is authorized to use a service before permitting
mechanism. it to do so.
Security Mode 4 introduced in Bluetooth 2.1 + Bluetooth does not provide for other security
EDR requires that encryption be used for all data traffic, services such as audit, integrity, and non-repudiation;
except for service discovery. As shown in Figure4, the these services should be implemented by other means if
encryption key provided to the encryption algorithm is they are needed.
produced using an internal key generator (KG). The KG
produces stream cipher keys based on the 128-bit link IV. SECURITY MODES IN BLUETOOTH DEVICES
key, which is a secret that is held in the Bluetooth
devices; a 128-bit random number (EN_RAND); and The family of Bluetooth BR/EDR/HS
the 96-bit ACO value. The ACO is produced during the specifications defines four security modes. Each
authentication procedure, The Bluetooth encryption Bluetooth device must operate in one of these modes,
procedure is based on a stream cipher, E0. A key stream called Security Modes 1 through 4. These modes dictate
output is exclusive-OR-ed with the payload bits and sent when a Bluetooth device initiates security,
to the receiving device. This key stream is produced
using a cryptographic algorithm based on linear A. Security Mode 1
feedback shift registers (LFSRs).12 The encryption Describes devices that have no capabilities for
function takes the following as inputs: the master device security authentication and encryption, leaving the
devices and connections vulnerable to attackers. These V. SECURITY THREATS IN BLUETOOTH DEVICES
devices do not prevent other Bluetooth-enabled devices
There are now billions of Bluetooth devices in use,
from establishing connections, and they can respond to
malicious security violations are common events now
security requests such as for authentication or
and it is expected to increase in the near future. On the
encryption that are initiated by a remote device. All
contrary, the increased usage of Bluetooth devices
v2.0 and earlier devices can support Security Mode 1,
makes security concerns even more alarming. Hence,
and v2.1 and later devices can use Security Mode 1 for
Bluetooth security architecture needs a constant
backward compatibility with older devices.
upgrading to prevent new unknown threats. Like any
B. Security Mode 2 other wireless communication system Bluetooth
transmission can be deliberately jammed or intercepted.
Is a service level-enforced security mode; security False or modified information could be passed to the
procedures may be initiated after link establishment but devices by the cyber criminals. Security threats in
before logical channel establishment. A local security Bluetooth can be divided into following major
manager, controls access to specific services. The
categories:
centralized security manager maintains policies for
access control and interfaces with other protocols and A. Disclosure Threat
device users. Varying security policies and trust levels
to restrict access can be defined for applications with The information can leak from the target system to
different security requirements operating in parallel. It an eavesdropper that is not authorized to access the
is possible to grant access to some services without information.
providing access to other services. This mode provides
for authorization to determine whether a specific device B. Integrity Threat
is allowed to have access to a specific service. The information can be deliberately altered to
Bluetooth service discovery can be performed prior to mislead the recipient.
any security challenges for authentication, encryption,
and/or authorization. The authentication and encryption C. Denial of Service Threat
mechanisms used for Security Mode 2 are implemented
in the controller. All v2.0 and earlier devices can The users can be blocked to get access to a service
support Security Mode 2, but v2.1 and later devices can by making it either unavailable or severely limiting its
only support it for backward compatibility with v2.0 or availability to an authorized user.
earlier devices.
D. Blue Bugging Threat
C. Security Mode 3 Blue bugging exploits a security flaw in the
Requires Bluetooth devices to initiate security firmware of some older Bluetooth devices to gain
procedures before the physical network link is fully access to the device and its commands. This attack uses
established. Bluetooth devices operating in Security the commands of the device without informing the user,
Mode 3 mandate authentication and encryption for all allowing the attacker to access data, place phone calls,
connections to and from the device. Service discovery eavesdrop on phone calls, send messages, and exploit
cannot be performed until after authentication, other services or features offered by the device.
encryption, and authorization have been performed.
Once a device has been authenticated, service-level E. Fuzzing Threat
authorization is not typically performed by a Security
Bluetooth fuzzing attacks consist of sending
Mode 3 device. All v2.0 and earlier devices can support
malformed or otherwise non-standard data to a device’s
Security Mode 3, but v2.1 and later devices can only
support it for backward compatibility purposes. Bluetooth radio and observing how the device reacts. If
a device’s operation is slowed or stopped by these
D. Security Mode 4 attacks, a serious vulnerability potentially exists in the
protocol stack.
Is a service level-enforced security mode in which
security procedures are initiated after physical and F. Secure Simple Pairing Threat
logical links are established. Security Mode 4 uses
Secure Simple Pairing (SSP), in which Elliptic Curve A number of techniques can force a remote device
Diffie-Hellman (ECDH) key agreement replaces legacy to use Just Works SSP and then exploit its lack of
key agreement for link key generation. The device MITM protection (e.g., the attack device claims that it
Authentication and encryption algorithms are identical has no input/output capabilities). Further, fixed
to the algorithms in Bluetooth v2.0 + EDR and earlier passkeys could allow an attacker to perform MITM
versions. Security requirements for services protected attacks
by Security Mode 4 must be classified as requiring Security threats like disclosure and integrity attacks
authenticated link key, requiring unauthenticated link typically compromise some sensitive
key, or requiring no security.
Information and therefore, can be very dangerous. alphanumeric) should be used for devices that require a
On the other hand, DoS attacks typically only annoy higher level of security.8
Bluetooth network users and are considered to be less
dangerous. C. Secure Simple Pairing
SSP was introduced in Bluetooth v2.1 + EDR for
VI. SOLUTION OF SECURITY THREATS use with Security Mode 4. SSP simplifies the pairing
The threats can be prevented by the following process by providing a number of association models
techniques: that are flexible in terms of device input/output
capability. SSP also improves security through the
A. Pairing and Link Key Generation addition of ECDH public key cryptography for
protection against passive eavesdropping and man-in-
Essential to the authentication and encryption the-middle attacks (MITM) during pairing.
mechanisms provided by Bluetooth is the generation of
a secret symmetric key, called the “link key.” Bluetooth
VII. CONCLUSION
BR/EDR performs pairing (i.e., link key generation) in
one of two ways. Security Modes 2 and 3 initiate link Bluetooth is a wireless technology which can do
key establishment via a method called Personal much more than just replace data cables between
Identification Number (PIN) Pairing (i.e., Legacy or devices. With the release of the Bluetooth version 4.0
Classic Pairing), while Security Mode 4 uses SSP. Both specification supporting higher data rates, greater range
methods are described below. and safer security measures, Bluetooth is clearly a good
choice for Wireless Networks. This paper shows a
B. PIN/Legacy Pairing unique way of utilizing this amazing technology to
For PIN/legacy pairing, two Bluetooth devices achieve efficient ways of communication. The use of
simultaneously derive link keys when the user(s) enter Bluetooth communication has always been at a personal
an identical secret PIN into one or both devices, level but never before at a public environment with high
depending on the configuration and device type. The user density. It has also been shown in this paper that a
PIN entry and key derivation are depicted conceptually careful design of such network can ensure effective
in Figure5. Note that if the PIN is less than 16 bytes, the security. Some limitations of this network have also
initiating device’s address (BD_ADDR) supplements been discussed which may have solutions in the future
the PIN value to generate the initialization key. The Ex through further ongoing research.
boxes represent encryption algorithms that are used
during the Bluetooth link key derivation processes. REFERENCES
[1] Jochen Schiller, “Mobile Communications” (second edition),
Addison Wesley Publications, 2003. pp. 290,292, 316
[2] C. Gehrmann, J. Persson, and B. Smeets, “Bluetooth Security”,
Computer Security Series, Artech House, 2004.
[3] Korzeniowski, Paul. 2005, “Bluetooth Security Threats Starting
to Sp read”, TechNewsWorld, February 02, 2005
[4] Paladugu, V., Cherukuru, N., & Pandula, S. (2001). Comparison
of security protocols for wireless communications.
[5] Slashdot. (2002, August 18). Wardriving from 1500ft Up.
[6] Stoneburner, G., Goguen, A., & Feringa, A. (2002, July). Risk
management guide for information technology systems. NIST
Special Publication 800-30.
[7] Wailgum, T. (2004, September 15). Living in wireless denial.
CIO Magazine.Jochen Schiller, “Mobile Communications”,
Second Edition, Addison Wesley Publications, 2003,
pp. 290-292.
[8] Bluetooth Version 4.0 Released. Bluetooth SIG, available
Fig. 5: Legacy Pairing at:http://www.bluetooth.com/Pages/High-Speed.aspx
[9] Keijo Haataja, “Security Threats and Countermeasures in
After link key generation is complete, the devices Bluetooth Enabled Systems”, Kuopio University Library, 2009,
complete pairing by mutually authenticating each other pp. 55-62
to verify they have the same link key. The PIN code [10] “The BlueBug”, a Bluetooth virus, available
used in Bluetooth pairing can vary between 1 and 16 at:http://trifinite.org/trifinite_stuff_bluebug.html
[11] John Oates, “Virus attacks mobiles via Bluetooth”, available
bytes of binary or, more commonly, alphanumeric at:http://www.theregister.co.uk/2004/06/15/ symbian_virus/
characters. The typical four-digit PIN may be sufficient [12] F-Secure Article on Lasco.A Worm, available at:http://www.f-
for low-risk situations; a longer PIN (e.g., 8-character secure.com/v-descs/lasco_a.shtml
Abstract—Transportation plays an important role in type of technology i.e., wireless ad-hoc network. This is
human life. If we look back towards the history of human a feasible system to improve the present traffic
being and development of civilization all it started after problem. Here we are going to connect a road side
the invention of wheel. Invention of wheel is first step for station with fast moving vehicles. This paper points out
making transportation easy. In this paper we also put
forward a technology which makes transportation more
about a prototype of WTPS. In it vehicles are equipped
reliable and safe and by using this system we can make an with wireless communication devices and considered as
auto-driving system. In this system there is no need of the moving sensors on the road. They can calculate the
manually driving for whole time however a driver can actual value of traffic speed, traffic density and other
take control when he wants. We can give a name for this traffic information parameter. This make further study
system wireless traffic positioning system (WTPS). This of communication subsystem of wireless traffic
Hi-tech system not only provides driver guidance message information service system focus on the end to end
but also internet services including text messages and delay, hop count, packet loss rate, throughput and
multimedia messages. In this system, vehicles are effective range of inter vehicle communication.
equipped with positioning device and they can be
considered as mobile sensors on road. Goods carriers are
The paper is organised as follows. In fragment II
an important part of transport through which humans can we are going to point out about some related work to
carry things at where they need. Generally super highway this paper that has been done before this paper. In
concept work for both light four wheeler and heavy fragment III guiding system for vehicle on
vehicles but as concerned for heavy to heavy vehicles they superhighways, in fragment IV we are going to present
are difficult to drive manually and if they are caring some a prototype of wireless traffic positioning system. In
valuable things then safety are a main challenging task. fragment V future research direction is given. In
Therefore in this paper we have tried to give a solution for fragment VI conclusion is given.
that. Further we can get an advantage of superhighway
that we can manage the traffic density at peak hours on
road by blocking the rush of vehicles on superhighways II. RELATED WORK
before entering. VANET work for short range communication
Keywords: Vehicles ad hoc network (VANET), Vehicle effectively by which we can avoid the collision of
to vehicle communication (V2V), vehicle to infrastructure vehicles by giving a threat against accidence acceptable
communication (V2I) distance. VANET is also favourable to auto payment of
toll tax and parking rent. Many works have been done
I. INTRODUCTION to alive this technology some of important works that
comes in our knowledge are given below.
Today developed and developing nations are
A project was done named “Application Level
facing a serious problem of traffic congestion. So it is
Protocol for Accident Reconstruction in Vanet” in
necessary to make and improve intelligent
university of south Carolina, in sep 2007.An another
transportation system in which vehicles are equipped
project was done in swiss fedral institute of technology
with specific kind of sensors that can sense position of
zurich in 2004, the title name was “Enginnering and
vehicles and transmit the real time information to its
Simulation of Mobile ADHOC Protocol for VANET on
centralize node. Mainly technology Goals are Safety on
Highways and in Cities”. Some other works are
Roads (huge list of situations/applications considered),
“Increasing Broadcast Relibility in vehicular AD-HOC
Reducing accidents, Reduce traffic congestion,
Networks”, in 2006 at university of Michigan by Nath
Alleviating accident damages, Traffic Conditions,
an Balon, a Masters project “Optimizing Dissemination
Improve transport efficiency, Monitor traffic demand
of Alarm Messsage in Vehicular ADHOC Networks” in
Environment, Reduce pollution, Driving Comfort,
2004 at university of Avignon france, an another
Driver assistance and other useful applications[4].
masters project “Provining Vanet Security through
At present the system which are use to collect
Position Verification” by Gyanesh Kumar Choudhary in
information of traffic are infrared detectors, ultrasonic
sep 2007at university of old dominion, a project on
detector video camera and Global Positioning System
“Security Issue in Vanet” in april 2010 Brac university,
but all of these systems are too much costly and much
Dhaka, Bangladesh,.A Sample theory of C. Lochert, B.
complicated. Therefore we have to focus on another
Sheuermann, M. Mauve. Probabilistic “Aggregation for vehicles and auto toll tax should be deposited
Data Dissemination in VANETs”, in 2007, and an electronically by e- banking technology (or by the
another industrial approach is done in Feb 2006 by plastic money). If any vehicles that are not registered
French motorways companies in which 51 partners are come into the superhighway, a warning message is
supporting (belongs to 12 different European countries), given to it and finally automatically it inform to police
the cost of this project is near to 38 million Rs. First station.
workshop was held in 2004 on vehicular adhoc
“Broadcast Reception Rates and Effects of Priority IV. RESEARCH ACHIEVEMENTS
Access in 802.11-Based Vehicular Ad-Hoc Networks”
Marc Torrent-Moreno, University of Karlsruhe Daniel Multiple access protocol use to coordinate access to
Jiang, Daimler Chrysler RTNA, Inc. Hannes a shared link. ALOHA is first developed random access
method, at the University of Hawaii in 1970. It is
Hartenstein, University of Karlsruhe. Some realistic
highway movement patterns (Courtesy of DC designed for LAN with a data rate 9600 bps. Since in
Germany), Realistic radio propagation model at random access technique, each node has right to use
shared link any time without control of other nodes.
highway scenarios using Nakagami distribution
(Courtesy of DC Palo Alto), Verified ns-2 802.11 MAC ALOHA has a problem that the packets of two nodes
and PHY layer (Uni-Karlsruhe) adjusted toreal 802.11a can be collide on shared link if they want to send
packets simultaneously. The maximum success rate of
Atheros Chips (values courtesy of Atheros).[5,6,7,8]
ALOHA protocol is 18.4 %for a fixed message length
[1].In 1972, S-ALOHA (slotted ALOHA) method was
III. GUIDING SYSTEM FOR VEHICLES ON
proposed by Roberts Basically S- ALOHA was a type
SUPERHIGHWAYS
of multiple access technique (discrete in time), in which
This system is based on vehicular ad-hoc network. whole time was divided into discrete intervals, each
Here all vehicles are connected via wireless link. So intervals corresponding to one frame. Hence it is
information of traffic is shared between all the vehicles. required to wait for the beginning of the next slot [9]. In
When a vehicle enters into the superhighway they have this probability of collision is just Pk=e-G(1-e-G)k-1 [9].
to register in superhighway network, this process is Therefore to overcome this problem, another technique
very much similar to a mobile cell phone when a cell was developed i.e. CSMA (carrier sense multiple access
phone goes outside from its home location it has to technique) that work on the principle of “Listen before
register in foreign network. If any vehicles that does not talk” means sense before transmission. Persistent and
want to recognise itself then road side station ask a non persistent are the two different strategies used in
question to unauthorized vehicle who are you? This CSMA. In case of persistent strategy, node Sense the
question is not as such actually road side station send a channel if it finds channel idle, transmits the frame and
coded digital message in repeatedly manner up to a if it finds channel busy sense carrier again.
limit. If vehicles give reply, that is registered and
system stores this information and assign a specific ID.
If that vehicle is not answering then that vehicle is
stopped by superhighway network and information will
be passed to traffic police. if a vehicle has register it has
to define its destination and its vehicle configuration
that is basically defined by the automatic device of the
vehicle that contains information about the maximum
speed of the vehicle, fuel tank information and many
other fundamental information that are required to drive
a vehicle on superhighway automatically. Now this
system can be compared with auto flying state mode on
airways in which a track is fixed and pre-computed.
This system is very useful for detection of stolen
vehicles.
Fig. (a): A Prototype of Superhighway
As shown in fig (a) many link roads are connected
with superhighway. Superhighway may be one way or In case of non persistent technique, node sense
two ways and there may be more than three lanes. channel if it finds idle send the frame and if it finds
Speed is pre fixed according to the better results. At busy it wait for random period of time then sense again.
near the junctions, minimum speed should be there and CSMA cannot remove the problem totally. Let us
as we go far to junction it increases and any one or two suppose when two nodes sense the medium no body
lane should be reserve for maximum speed of transfer any packets so both nodes finds shared link
superhighway by using that lane people can save their idle. Therefore both nodes will transfer their packets but
time. Before entering on superhighway, registration of they create interference. Hence CSMA can reduce the
possibilities of collision but it cannot eliminate it. FDMA and receiver send a idle message to its
CSMA/CD (carrier sense multiple access with collision neighbour. In this protocol data signal and
detection) is another techniques that helps to overcome acknowledgement are transmitted through a different
the problem of collisions. First of all it sets a back off technique in which first of all FDMA technique is used
limit and uses persistence strategy and send frame if and further TDMA techique is used for same carrier
nodes does not get acknowledgement it means there is frequency bands. 3D fig(c) is shown this multiple
collision it sends a jam signal and increase a back off access technique in which whole frequency channel is
limit count if back off limit count exceeds it abort. If it devided into many small frequency band and further
does not exceed it wait for back off time and further use can be shared by many ad hoc user by dividing whole
persistence strategy to sense the shared link. There are time into small time slots. As shown in 3D fig(c), third
many other protocol like CSMA/CA that give an idea of axis repersent the power of signal.
channel access for multiple user.
Here we are using a new type of medium access
protocol which knows better to superhighway as
compare to any pre-existed protocol. The name of this
protocol is NMAPS (new multiple access protocol for
superhighway). In this protocol, we are using different
techniques TDMA, FDMA & CDMA for
communication among the nodes and road side station.
Both TDMA & FDMA is for sending the data from
node to another node (V2V) and from node to road side
station. However CDMA is for signalling that tells to its Fig. (b)
neighbour node that this node is either transmitting or
Power is an important parameter because it shows
receiving data. As we know, main problem of adhoc
the ability of signal how it can go far. For request
network is fast changing topology of nodes and to get
message and reply message, it use CDMA technique, in
new paths for adhoc network and as far as concerned to
which transmitt power is reduced to less than of
vehicular adhoc network vehicles moving with very fast
interference limit however the power is enough to
speed. Therefore network topology changing
maintain the required signal to noise ratio for
probability is high on normal city highway but here (on
satisfactory communication. Since a CDMA signal is
superhighway) this is not a big issue because we can
transmitted on existed frequency band Hence it is
adjust vehicles as we want (vehicles are being drive
treated as noise for other demodultion techniques (Since
automatically). As shown in fig whenever a node wants
power level of CDMA signal is kept less than 20% of
to transmitt a data packet. It has to send a CDMA
other modulated signals like AM & FM) and due to
transmission request message towards the destination. If
this, CDMA signal is not disturbed to data and
any by pass node comes in the route of this request
acknowledge signal. Maximun capacity can be achieved
message that helps to forward this message towards
when signal to noise ratio of every signal is kept at the
destination. As this message comes on destination node.
minimum level.
Status of destination node (receiver node) is checked
(However status of node will determine by a prior node
to final destination) if it finds busy then further
communication is restricted and a CDMA reply send to
transmitter node and as this message arrive at
transmitter node it generate a idle mssage that sends to
its neighbour nodes.After a random period of time it
transmit again a CDMA request message to destination
node (after sending a CDMA transmitting message to
its neighbour). If destination node finds idle. It send a
different CDMA reply message towards transmitter Fig. (c): 3D Representation of Power Divides with Frequency
as Well as Time
node. As this CDMA reply message arrive at
transmitter node. It starts to send data packets towards
V. FUTURE RESEARCH DIRECTIONS
destination and wait for acknowledegement upto a limit
beyond this limit it transmit again a duplicate copy of In future we can make a software model of this
this message. After complition of communication both research work and further more we will study about the
transmitter and receiver both transmitter and receiver end to end delay, hop count, packet loss and other
send a idle message to its neighbour. In this protocol communication parameters for equipped vehicles. There
data signal and acknowledgement are transmitted are various software available that can help this
through a different technique in which first of all research Opnet, Qualnet, NS2, NS3 and Matlab etc.
Abstract—Botnet has recently been recognized as fraud, identity theft and information exhilaration is
one of the most significant security threats/worms of the main hazardous behavior which is associated with the
Internet. Latest attacks are increasingly complex, and botnet. Botnets apply a self-propagating function to
utilize many strategies in order to perform their intended infected hosts. Given the importance of the problem,
malicious/hazardous task. Attackers have developed the
ability of controlling vast area of infected hosts,
significant research effort has been invested, to gain a
characterized by complex executable command set, each better understanding of the botnet phenomenon.
involved in cooperative and coordinated attacks. So we Botnet detection: It is the technique of detecting a
will propose the advanced approach related botnet bots from the network. Botnet detection strategies are
detection and analysis in the near future. This paper broadly divided into two types:
demonstrates a novel approach of botnet investigations • Host based approach: Detecting bots
and defense mechanisms. activities on a single machine.
Keywords: Bot, Botnet, C&C, Analysis • Network based approach: Detecting bots
activities on a network.
I. INTRODUCTION One approach to study botnets is to perform
passive analysis of secondary effects that are caused by
Currently, Computer networks and hosts have the activity of compromised machines [2]. For
always been under attack by computer-generated example, researchers have collected spam mails that
problems. Bot, malicious software has been a global were likely sent by bots. It is collecting the data
crisis. A network of bots constitutes a botnet which is a
through monitoring activities which can be tracked
potent general purpose distributed supercomputer.
without interfering with the environment or tampering
Botnet represent a very serious threat to the Internet
with the evidence. Other researchers analyzed IRC
security because they can be used to initiate massive
attacks against which there are no effective mitigation traffic, capable of identifying botnet related activities.
strategies. The Botnet is a network of large number of A more active approach to study botnets is via
the infected end host called Bots which is controlled by infiltration. It contains approaches that involve
a remote human operator called “Botmaster”. A bot is interaction with the information sources being
installed in a compromised machine and botmaster monitored. Infiltration of botnets can be divided into:
operates the bots via command and control(C&C). The software and hardware based techniques. The first
term ‘bot’ is used to denote a computer that is infected covers research on the bot executable and monitored
by malicious code which often exploits software traffic to achieve control and conduct measurements.
vulnerabilities on the computer to allow a malicious The latter can be applied if access to the command-and-
get-together commonly denoted as ‘botherder’ to control server is possible and may be used to wiretap
control the computer from a remote location without the the communication. According to the command and
user’s knowledge and consent [1]. control (C&C) models, botnets are separated into two
Once infected with a bot, the victim host will join a groups, centralized (e.g., IRC and HTTP) and
botnet, which is a network of compromised machines distributed (e.g., P2P). Centralized botnet utilize two
that are under the control of a malicious entity, mechanisms to get the command from the server,
typically referred to as the botmaster. Botnets are the which is push and pull. In the push system, bots are
primary means for cyber-criminals to carry out their associated to the C&C server (e.g., IRC server) and
nefarious tasks, such as sending spam mails, launching wait for the commands from the botmaster. In
denial-of-service attacks, or stealing personal data such contrast, in the pull mechanism, the botmaster sets the
as mail accounts or bank credentials. This reflects commands in a file at C&C server (e.g., HTTP server),
the shift from an environment in which malware was
and the bot often connect to the server to read the most
developed for fun, to the current situation, where
recent commands. While in centralized structure all
malware is spread for financial profit. Denial-of-service
bots receive the commands from a definite server, in
(DoS) attacks, phishing, spamming, key logging, click
automated propagation vectors. For example, The A P2P botnet do not requires formal coordination
Slammer worm used a single vulnerability to infect and even if a node is taken offline by the defender, the
hosts while more modern bots have many distinct, network still remains under the control of the attacker.
completely automated propagation vectors/ SDBot A botmaster transfer command to a bot peer, a
(rBot) propagates using a number of different command spreads all zombie peers by communicating
mechanisms including open files shares, P2P networks, with each other. They have the advantage of being more
backdoors left by previous worms, and using exploits of difficult to destabilize as they do not have a unique core
numerous common Windows vulnerabilities. which issues orders and/or sharing resources and
information, making use of the facilities of traditional
B. C& C (Command and Control) P2P networks allow a high connection and
Communication is the next major problem of the disconnection ratios. Each node has greater structural
botnet attacker. Most attackers would communicate to complexity because all of them can act as both, client
bots but do not interact to the exposed bots. Botnet has and server, being more difficult to intercept and study.
three potential topologies: P2P botnets aim at removing the failure point
Centralized: which is the main limitation and vulnerability of
A centralized topology is characterized by a central centralized networks [6,7] P2P communication system is
point that communicates among clients. Malicious IRC much strong, complex and does not guarantees message
botnet/ HTTP botnet is typical Centralized botnet [7]. delivery or latency. Transferring command of P2P
They can be easily detected as many Clients connect the botnet is slow to compare with centralized botnet.
same point, and it can stop the activity of botnet with This means that the compromise of a single bot does
blocking C&C. Usually consisting of a central node not necessarily mean the loss of the entire botnet. e.g.,
distributing messages between network clients. They Peacomm, Mega-D, Waledac etc.
are characterized by:
• Low latency due to the small number of hops
required to transmit the orders from the
botmaster.
• Direct connection to order distribution nodes,
which can compromise the security of the
network in case of accidental detection of a
node.
• Implemented using different communication
protocols, most typically IRC and HTPP.
2) Hybrid
A hybrid peer-to-peer botnet based on the
unstructured P2P protocols[8]. A hybrid botnet will be
Fig. 2: Centralized Botnet Structure divided into servent and client bot. The servent bot
receives the commands from the bot master, and
1) Peer-to-peer forwards it to the client. Example is the Nugache
P2P topology does not have Command & Control botnet (2006). The hybrid P2P botnet is equivalent to a
server. In a peer to peer architecture a node can act as C&C botnet where servent bots take the role of C&C
both client and server and there is lack of centralized servers, the number of C&C servers (servent bots) is
point for command and control. greatly enlarged, and they interconnect with each other.
In hybrid P2P Botnet, in comparision to current botnet,
it is harder to shut down,monitored and hijacked.
Li et al. [13] focused on all phases of botnets as recruiting new bots and social engineering tricks to
Spreading, Forming, Waiting for Commands. Network spread malware binaries. Hydra fast-flux network, use
traffic and honeypots is novel approach to monitor and of SQL injection attack tool is advanced features of
analysis of P2P lifetime. The network traffic monitoring Asprox. This paper introduced the botnet advanced
and analysis approach is useful to identify the existence features such as hydra fast-flux network.
of botnet in the networks. Kademlia, a new botnet AsSadhan et al. [15] discovered a bots by looking
protocol is designed which is a realistic method to deal for periodic behavior in the Command and Control
with index poisoning in this protocol. communication traffic in the monitored network. They
The new botnet over this protocol is called Overbot. extract from a bot’s Command and Control
Simulation tests were operated to evaluate the communication traffic the packet count sequence to
performance of index poisoning attack. To simulate a evaluate its periodogram.
P2P botnet with 4000 nodes, time in the simulated Riccardi et al. [16] proposed the framework aimed
network is moving 60 times faster than normal, by to manage the whole workflow of identifying,
using index poisoning technique. It is at least more analyzing, and mitigating a financial botnet, ranging
effectively to mitigate P2P botnets than before. from an initial malware analysis to the creation of
Dae-il et al. [7] studied the malicious HTTP2P specific feedback and knowledge shared with interested
botnet. Malicious botnet is evolving very quickly and parties cooperating to fight cybercrime.
use many ways to evade detection system. The change This paper presents a work-in-progress research
of protocol is the most important part of the malicious aimed at creating a system able to mitigate the financial
botnet’s evolution and evasion techniques. The initial botnet problem. The proposed system is based on a
malicious botnet was using the IRC protocol for novel architecture that has been validated by one of the
communication between the command and control biggest savings banks in Spain.
server and the zombie system. After that they used Goel et al. [17] focused on a practical host-based
HTTP protocol on the firewall-friendly and the P2P methodology to the collect evidentiary information from
protocol to escape Client/Server architecture. a Bot- infected machine. They studied on the analysis of
Investigation of waledac was to find out more about the BotNet behaviors, propagation, and method to detect
actual size of botnet. and stop their proliferation. There approach collects
Igor Kotenko et al. [14] proposed an approach of digital traces from both the network and physical
botnet simulation and defence in the Internet. It is memory of the infected local host, and correlates the
examined by interaction of different agents teams. information to identify the resident BotNet malware
Environment for the agent-oriented simulation was involved. Bot-herders are constantly evolving and
developed on the basis of OMNeT++ INET Framework. adapting the structure of their BotNets at the network
Botnet defence can be considered in two main level to create ever more robust control mechanisms,
categories botnet detection/response techniques and and to avoid current detection techniques.
measurement. Agent- based simulation of cyber attacks Binsalleeh et al. [18] focused on reverse
and cyber–defense mechanisms which combines engineering results for the Zeus crimeware toolkit, one
discrete-event simulation, multi-agent approach and of the recent and powerful crime ware tools that
packet- level simulation of network protocols. The emerged in the Internet underground community to
investigation of attack and defence scenarios has been control botnets. The Zeus crimeware toolkit is an
finished on the basis of analysis of two main classes of advanced tool used to generate very effective malware
parameter: the amount of incoming attack traffic before that facilitates criminal activities. They have also
and after the filter team, which network is the attack designed a tool to automate the recovery of the
victim, false positive and false negative rates of attack encryption key and the extraction of the configuration
detection. This paper shows the best result on blocking information from the binary bot executables. Analyze
the attack traffic. This software simulation environment and extract the Zeus C&C servers use encryption input
has been used to investigate various cooperative system.
distributed defence mechanisms. Flaglien et al. [19] proposed on the design
Borgaonkar et al. [1] studied the design and framework for discovering and correlating evidence
structure of the Asprox botnet. They investigated the C from multiple components in a particular computer.
& C structure used by Asprox botnet, the To combine multiple evidence sources, all the data
communication protocols, the drive-by download must be represented in the same format, designed a
technique for spreading malicious content and the common representation format for evidence from
advanced fast-flux service network. The main features multiple sources. The correlation method supports
of the Asprox botnet are the use of centralized forensic investigations using link mining techniques.
command and control structure, HTTP based Stone-Gross et al. [2] focused on efforts to control
communication, use of advanced double fast-flux of the Torpig botnet. It is large and targets a variety of
service networks, use of SQL injection attacks for applications, gathers a rich and diverse set of data from
the infected victims. Analyzing a cyber-warfare strategies may be implemented directly on the network
scenario, the possibility of taking over botnet control without an overall solution to all hosts on the Internet.
may also be considered to conduct counter- attack Containment mechanisms must be considered in three
actions. They describe their experience in actively aspects as follows: Detection and reaction time,
seizing control of the Torpig botnet for ten days. Torpig Strategy used to identify and contain Solution topology
introduces Mebroot only when necessary to understand and scope. Two techniques were used to gather such
Torpig’s behavior. These primarily focus on the Master information:
Boot Record (MBR) overwriting rootkit technique mwcollectd developed a software, honeypot [21].
employed by Mebroot. The honeypots compromised and joined a botnet.
Balzarotti et al. [20] focused on infection data Behaving as normal bots in the botnet, these honey pot
from major botnets: Conficker, MegaD, and Srizbi. In spies provide valuable information about the monitored
this analysis, they examine commonly-infected botnet activities. The software was designed to be
networks which appear to be extremely prone to infected and automate the information gathering tasks.
malware infection. They provide an in- depth passive It acts like a low-interact-tive honeypot, emulating
and active measurement study. CAIDA provides basic services and vulnerabilities. Such emulations increase
information about the victim distribution of the the likelihood that the botmaster would realize that it is
Conficker botnet in terms of their IP address space and a monitoring device and evade the system. After botnet
physical location. They have proposed and verified infiltration, the authors propose a way to turn the botnet
cross-botnet prediction techniques to predict unknown off, changing the DNS server configuration. If the name
victims of one botnet from the information of the other
resolution is redirected to private addresses, bots could
botnet if they have similar infection vectors. In future
not connect to the C&C server anymore, and the
work, they will study new approaches to explain
botnet itself would be disarticulated. This approach
relationships between geopolitical locations and
requires the cooperation of the DNS provider. With the
malware infection more clearly with some realistic
examples. involvement of the DNS provider, one may also take
control of the entire network by simply configuring a
IV. BOTNET DEFENSE TECHNIHUES C&C server. Network forensics involves monitoring
network traffic and determining if there is an anomaly
Defense techniques against bots focused on two in the traffic and ascertaining whether it indicates an
main activities: propagation and bot communication. attack. If an attack is detected, then the nature of the
Combating bot/worm spread directly affects the attack is also determined. Botnet forensic techniques
number of compromised machines in a botnet, thus enable investigators to track back the attackers. The
reducing the network power and utility to the ultimate goal is to provide sufficient evidence to allow
botmaster. Conversely, considering already infected the perpetrator to be prosecuted. Botnet forensics can
machines, another form of defense is to stop the be used to analyze how the attack occurred, who was
communication between bots and C&C servers. The involved in that attack, duration of the exploit, and the
two defense approaches mentioned in main areas:
methodology used in the attack. Botnet security
prevention, treatment and containment. For the
protects system against attack while botnet forensics
defense to be effective there must be actions
focuses on recording evidence of the attack. Botnet
performed by users, network administrators, and
security products are generalized and look for possible
Internet Service Providers (ISPs). For propagation, the
most important features for worm spreading are the harmful behaviors. This monitoring is a Botnet process
number of vulnerable machines, infection duration, and and is performed all through the day. But, network
infection rate. Prevention techniques thus aimed to forensics involves postmortem investigation of the
reduce the vulnerable population, limit the worm attack and is initiated notitia criminis (after crime
spread, and reduce the botnet size. Preventive actions notification). It is case specific as each crime scenario
are related to secure software development, updated is different and the process is time bound. Botnet
system maintenance, vulnerability removal, antivirus security as the data for forensic analysis are collected
program use, training, and user social economic power. from security products like firewalls and intrusion
Mechanisms employed in malware control are also detection systems. The results of this data analysis are
effective in combating botnets, but that action alone is utilized for investigating the attacks. However, there
not enough to stop these threats. Treatment is related to may be certain crimes which do not breach. Botnet
disinfecting zombies to reduce the number of bots and security policies may be legally prosecutable. These
perform system updates to reduce vulnerable hosts and crimes can be handled only by botnet forensics, so
the worm spread rate. The main reasons for using botnet defences can be organized by phase (Avoidance,
containment and blocking strategies include the detection, and reply) and by role (agent or target).
following: Blocking can be automated after bot Defenses for each of the resulting six situations (three
detection, without depending on human action. These phases and two roles for each) are presented in Table I
and tactics that can be used against the problem. [10] M.M. Masud, T. Al-khateeb, L. Khan, B. Thuraisingham, and
K. W. Hamlen, "Flow-based identification of Botnet traffic by
Particularly, finding ways to reduce the demand mining multiple log files," in Distributed Framework and
element could result in less use of botnets in whole Applications, 2008. DFmA 2008. First International
classes of behavior. Future work is related to Conference on, 2008, pp. 200-206.
comprehensive analysis of cooperation effectiveness of [11] P. Kalakota and C. T. Huang, "On the benefits of early filtering
of botnet unwanted traffic," in Computer Communications and
various attack and defense teams and inter-team Networks, 2009. ICCCN 2009. Proceedings of 18th
interaction, the implementation of adaptation and self- Internatonal Conference on, 2009, pp. 1-6.
learning defence to protect against manipulation by [12] R. Li, L. Gan, and Y. Jia, "Propagation model for botnet based
attackers, the expansion of attack and defence library, on conficker monitoring," in Information Science and
Engineering (ISISE), 2009 Second International Symposium
and the investigation of new defence mechanisms. The on, 2009, pp. 185-190.
important part of future research is providing numerous [13] X.-n. Li, Y. Liu, and H. Zheng, "Peer-to-Peer botnets:
experiments to study the effectiveness of prospective Analysis and defense," in Communication Software and
defence mechanisms against botnets. Networks (ICCSN), 2011 IEEE 3rd International Conference
on, 2009, pp. 140-143
[14] M. Essaaidi, M. Malgeri, C. Badica, I. Kotenko, A.
REFERENCES Konovalov, and A. Shorov, "Simulation of Botnets: Agent-
Based Approach," in Intelligent Distributed Computing IV.
[1] R. Borgaonkar, "An Analysis of the Asprox Botnet," in
vol. 315: Springer Berlin Heidelberg, 2010, pp. 247-252.
Emerging Security Information Systems and Technologies
[15] B. AsSadhan, J. M. F. Moura, and D. Lapsley, "Periodic
(SECURWARE), 2010 Fourth International Conference., 2010,
Behavior in Botnet Command and Control Channels Traffic," in
pp. 148-153.
Global Telecommunications Conference, 2009. GLOBECOM
[2] S.-G. Brett, C. Marco, C. Lorenzo, G. Bob, S. Martin, K.
2009. IEEE, 2009, pp. 1-6.
Richard, K. Christopher, and V. Giovanni, "Your botnet is my
[16] M. Riccardi, D. Oro, J. Luna, M. Cremonini, and M. Vilanova,
botnet: analysis of a botnet takeover," in Proceedings of the
"A framework for financial botnet analysis," in Crime
16th ACM conference on Computer and communications
Researchers Summit (eCrime), 2010, 2010, pp. 1-7.
security Chicago, Illinois, USA: ACM, 2009.
[17] S. Goel, F.W. Law, K.P. Chow, P.Y. Lai, and H.S. Tse, "A
[3] L. Wei, M. Tavallaee, G. Rammidi, and A. A. Ghorbani,
Host-Based Approach to BotNet Investigation?," in
"BotCop: An Online Botnet Traffic Classifier," in
Digital Forensics and Cyber Crime. vol. 31: Springer Berlin
Communication Networks and Services Research Conference,
Heidelberg, 2010, pp. 161-170.
2009. CNSR '09. Seventh Annual, 2009, pp. 70-77.
[18] H. Binsalleeh, T. Ormerod, A. Boukhtouta, P. Sinha, A.
[4] E.S. Pilli, R.C. Joshi, and R. Niyogi, "Network forensic
Youssef, M. Debbabi, and L. Wang, "On the analysis of the
frameworks: Survey and research challenges," Digital
Zeus botnet crimeware toolkit," in Privacy Security and
Investigation, vol. 7, pp. 14-27, 2010.
Trust (PST), 2010 Eighth Annual International Conference on,
[5] S. r. S.C. Silva, R. M. P. Silva, R. C. G. Pinto, and R. M. Salles,
2010, pp. 31-38.
"Botnets: A survey," Computer Networks, 2012.
[19] K.F. a. A.A. Anders Flaglien, "Identifying Malware Using
[6] M. Bailey, E. Cooke, F. Jahanian, X. Yunjing, and M. Karir,
Cross-Evidence Correlation," Advances in Digital Forensics,
"A Homeland Security, 2009. CATCH '09. Cybersecurity
vol. VII, pp. 169-182, 2011.
Applications & Technology, 2009, pp. 299-304.
[20] R. Sommer, D. Balzarotti, G. Maier, S. Shin, R. Lin, and G.
[7] J. Dae-il, K. Minsoo, J. Hyun-chul, and N. Bong-Nam,
Gu, "Cross-Analysis of Botnet Victims: New Insights and
"Analysis of HTTP2P botnet: case study waledac," in
Implications,"in Recent Advances in Intrusion Detection. vol.
Communications (MICC), 2009 IEEE 9th Malaysia
6961: Springer Berlin Heidelberg, 2011, pp. 242-261.
International Conference on, 2009, pp. 409-412.
[21] S. Vimercati, P. Syverson, D. Gollmann, F. Freiling, T. Holz,
[8] D. Dong, Y. Wu, L. He, G. Huang, and G. Wu, "Deep analysis
and G. Wicherski, "Botnet Tracking: Exploring a Root-Cause
of intending peer-to-peer botnet," in Grid and Cooperative
Methodology to Prevent Distributed Denial-of-Service Attacks,"
Computing, 2008. GCC'08. Seventh International Conference
in Computer Security –ESORICS 2005. vol. 3679: Springer
on, 2008, pp. 407-411.
Berlin Heidelberg, 2005, pp. 319-335.
[9] Y. Kugisaki, Y. Kasahara, Y. Hori, and K. Sakurai, "Bot
Detection Based on Traffic Analysis," in Intelligent
Pervasive Computing, 2007. IPC. The 2007 International
Conference on, 2007, pp. 303-306.
Abstract: Wireless network are going to be pervasive, bandwidth utilization. If two end hosts can obtain
and hence it is almost impossible to keep ourselves away network information from routers, they will be able to
from this technology, but still wire network has its own
execute congestion control more efficiently. With
significance which can not be ignored due its reliability.
The major difference between both technologies is that respective to a path, congestion usually occurs in the
wireless has more error prone attitude to packet bottleneck point which has the minimum available
transmission with respect to wired network. Among the bandwidth. If the sender can adjust the data rate
available transport level protocol, TCP is the most dynamically according to the status of the bottleneck,
prominent due to its characteristics and lay term use in TCP induced congestion should be avoid or dissolved
application. Rather than concentrating on new transport efficiently. Therefore, Yao-Nan Lien et al. [11] propose
layer protocol, we have improved the present TCP
congestion control mechanism so that it can be used
a new congestion control mechanism TCP Muzha,
universally on wired and wireless technology. One of the which takes router-assisted approach to dynamically
reasons behind the degradation of TCP congestion control adjust the size of congestion window (i.e. data rate)
mechanism performance is considering packet loss due to according the router information, and to deal with
corruption in channel as network congestion. In this paper random loss.
we have discussed the TCP congestion control mechanisms
in Wireless Network. B. Congestion Control in Wireless Networks
Keywords: Modified TCP, TCP congestion control, Adaptive Window based control as used by TCP
Adaptive window management. [2] was designed for use in wired networks. It becomes
imperative to ask whether the same mechanism will
I. INTRODUCTION AND PROBLEM STATEMENT also work in the case of wireless networks. In order to
A. TCP Congestion Control be able to answer this question, we need to understand
the differences between wired and wireless networks.
Transmission Control Protocol (TCP) [9, 10] used The most dominant difference between wired and
in wired networks uses Adaptive Window Management wireless networks is the time- varying random
to control packets being offered to the network. This fluctuations in the wireless channel. Hence, it is not
algorithm allows the sender to adapt its sending rate to clear whether mechanisms being used in wired
the current traffic conditions in the path it is using for networks will perform comparably in wireless
the particular flow. The TCP protocol [3] incorporates networks as well. In TCP, the window size W is
sender-receiver flow control through reduced whenever packets are lost assuming that the
acknowledgements being sent, from the destination to loss is due to traffic congestion along the path of the
the source node, for the packets received. Using these flow under consideration. This assumption is not valid
acknowledgement packets, the congestion control in wireless networks since the channel may have been
mechanism operates as follows: The source TCP uses a in a poor state thereby causing the packet to be lost
Window parameter which is the maximum number of at the physical layer (OSI Model) itself. The statistics
unacknowledged packets that it can have at any given of this type of loss may be very different from a packet
time during the connection. This restricts the maximum dropping at a router queue due to traffic congestion [6].
number of packets that the sender can offer to the Thus, we need to understand:
• Whether there is a need to distinguish between
network without having received the acknowledgement
traffic related losses and channel related
for any of them. Adaptive Window Control involves
losses in wireless networks.
adaptively changing the size of the transmission
• If we agree to distinguish between the two,
window based on the acknowledgement packets that
we need to find efficient ways of achieving
the source has received [6]. this objective.
Most of the TCP are not aware of network
condition such that they may not be able to control
congestion precisely and promptly resulting in unstable
• Formulate strategies to take different link. Wireless networks are characterized by large
corrective actions for each of the above error rates due to fading, noise, interference from
causes. other sources and mobile host movement. Due to large
Mukesh Kumar Dhariwal et al. [7] presented an error rate, packet losses due to congestion in the
improved mechanism for TCP congestion control. It wireless link are more as compared to congestion
considers the influences sending rate to TCP sender’s losses. The packet loss rate depends on lost packet it
packet only by the congestion. The drop rate for TCPM will be occur due to congestion. Packet drop rate is
is calculated according to total packet received and total not adjusted when the congestion loss rate is low, but it
packet acknowledgment in network. So, there is less is required to decrease the drop rate rapidly to
packet drop in transmission. improve the reliability when the congestion loss rate
becomes higher. Otherwise, there will be more lost
Rejaie et al. present in [9] an adaptation scheme
packets due to congestion and more packets will be
called the rate adaptation protocol (RAP). Just as with
retransmission, responding the poor transmission
TCP, sent packets are acknowledged by the receivers
reliability and more energy consumption of mobile
with losses indicated either by gaps in the sequence
hosts. Total drop rate is denoted by Td [8]. If sender
numbers of the acknowledged packets or timeouts. The sends n packets in a time period T and m packets of
sender estimates the round trip delay using the them be discard because one or more bits error caused
acknowledgment packets. If no losses were detected, by wireless link congestion, the total drop rate is define
the sender periodically increases its transmission rate as
additively as a function of the estimated round trip Td = Tr–Ta
delay. After detecting a loss the rate is multiplicatively
reduced by half in a similar manner to TCP. II. MODIFIED TRANSPORT CONTROL PROTOCOL
C. Problem Identification Modified Transport Control Protocol (TCPM) is
the improved TCP congestion control mechanism for
Unfortunately present TCP mechanism [11, 10] wireless network [7]. It considers the influences to TCP
could not give expected result due to packet loss due to sender’s packet sending rate only by the congestion
corruption in channel. We know internet technology is but also by the link failure. It is a good reference to
changing fast and wireless network is becoming apply the TCP to wireless networks.
pervasive all around, and packet lost in wireless
network may not imply congestion in network because A. Fast Retransmission and Fast Recovery
it may happen due to weather condition, obstacles and If the network congestion or heavy corruption, the
multipath interfaces, mobility of wireless end-devices fast recovery algorithm will be taken. When the
signal attenuation and fading. So our assumption upon network congestion, set ssthresh to one-half the flight
which network congestion window has been changing size or double of MSS (maximum segment size)
in TCP congestion control mechanism is not window.
appropriate for wireless. The overall goal of congestion
control is to optimize the performance in a B. Congestion Avoidance Algorithm
communication network. This optimization means,
If the congestion window size (cwnd) is less than
roughly that sending rates at the data source should be
or equal to the slow start threshold size (ssthresh),
as high as possible, without overloading the network. TCPM is in slow start; otherwise TCPM is performing
The primary measure of network overload is packet congestion avoidance.
losses
When the arrival rate at a link exceeds capacity the C. Slow-Start Algorithm
corresponding queue start to build up, and when the
The slow start algorithm is used to start a
queue is full, packet must be discarded the bottleneck
connection of TCPM and the periods after the value
links in the network should be fully utilized. The
of retransmission timer exceed the RTO
requirement of a small loss rate implies that the
(retransmission timeout). In the start of TCPM, the size
average arrivals rate at each bottleneck link should of cwnd will be initialized to 1.
either match the link capacity exactly, or be very
slightly larger. D. Initial Window
III. VARIOUS
A PERFO
ORMANCE MATTRICES USED Fig 1 & 2 shows
s signal received wiith error of
IN TCCP CONGESTIO
ON CONTROL MECHANISM [7] diffeerent variants with pause tiime and numbber of node
in network
n [6]. It is observeed that TCPM M has less
a) Paccket loss is thhe measure of total discaarded
errorr signal thann other TCP P variants, because
b of
packket due to corruption or duue to packet drop.
optimmal routing path betweeen sender an nd receiver.
It caan be calculaate by subtraccting total received
Increement in thee no. of noodes causes congestion
packkets by serverr with total sennt packet by client.
c
conssequently signnal were distorrted.
b) Signnal received with error is i the measuure of
signnal received, but they havve error. The error
mayy be occurringg due to noisse or due to heavy h
trafffic.
c) Throoughput is thhe measure of o the number of
packkets successffully transmittted to their final
desttination per unnit time. It iss the ratio bettween
the numbers
n of seent packets vss. received pacckets.
d) Bytees received are
a the measuure of total packet
p
receeived by servver. The packkets may be drop
due to heavy traaffic. So received packetss may
be vary
v accordingg to traffic connditions.
Fig. 1
V. CONCLU
USION
186 Bhag
gwant Institute of
o Technology, Muzaffarnagar,
M Uttar Pradesh, India
Study on TCP Congestion Control Mechanism in Wireless Network
mobile adhoc network. The simulation shows that the [4] H. Balakrishna, V.N. Padmanabhan, S. Srinivasan, and R.H.
Katz, A Comparison of Mechanisms for Improving TCP
performance of TCPM is better than other TCP Performance over Wireless Links, ACM SIGCOMM, August
variants. The performance of TCP variants are analyzed 1996.
on various parameter like pause time, size of network. [5] S.H. Low, F. Paganini, and J.C. Doyle Internet Congestion
Result shows that the performance of TCPM is better in Control, IEEE Control Systems Magazine, Feb. 2002.
[6] Umesh D.Nimbhorkar, Mohit Garg, “Congestion control in
high density nodes as in this circumstances sender can wireless networks”, May 5, 2004.
get different paths through different nodes. [7] Mukesh Kumar Dhariwal, Sanjeev Sharma,” An improved
mechanism for congestion control in TCP for Adhoc
REFERENCES network”,IJCA(0975-8887) Volume 20, April 2011.
[8] Xu Chang-Biao, Long Ke-Ping, Yang Shi Zhong. Corruption-
[1] G. Holland and N. Vaidya, “Analysis of TCP performance over based TCP rate adjustment in wireless networks. In: Chinese
mobile ad hoc networks,” in Proc. ACM/IEEE Int. Conf. on Journal of Computers, 2002, 25(4): pp.438-444.
Mobile Computing, Seattle, WA, USA, Sept. 1999, [9] R. Rejaie, M. Handley, and D. Estrin, “RAP: An end-to- end
pp. 219–230. rate-based congestion control mechanism form real time streams
[2] W.G. Zeng and Lj. Trajkovic, “TCP packet control for wireless in the internet,” in INFOCOM (3), 1999, pp. 1337–1345.
networks,” in Proc. IEEE Int. Conf. on Wireless and Mobile [10] Vicente E. Mujica V., Dorgham Sisalem, Vicente E. Mujica
Computing, Networking and Communications (WiMob 2005), V., Dorgham Sisalem, “TCP-Friendly Congestion Control over
Montreal, Canada, Aug. 2005, vol. 2, pp. 196–203. Wireless Networks”, in Proc. of European Wireless 2004,
[3] J. Postel, “Transmission control protocol,” RFC 793, Barcelona, Spain, February 2004.
Sept.1981. [11] Yao-Nan Lien and Ho-Cheng Hsiao, “A New TCP Congestion
Control Mechanism over Wireless Ad Hoc Networks by Router-
Assisted Approach”.
Abstract—Maintaining Network Security has become intrusion detection process [13], leveraging existing
an important issue in various government and Industrial RDBMS infrastructure can be both efficient and
sectors. Many Intruders involve to a great extent in the effective. The current paper presents slight change in
Computing Environment. Modern Intrusion detection the DAID
systems need to include various important characterstics
suc has reliability, extensibility, manageability, economic.
(Database-centric Architecture for Intrusion
Inspite of established intrusion detection systems based on Detection) which can be used for all types of databases,
data mining, still some issues exist in design and including Oracle Database. This RDBMS-centric
implementation. The Paper describes a Database Centric framework can be used to build, manage, deploy, score
Architecture that takes the advantage of Data Mining with and analyze data mining-based intrusion detection
RDBMS. By implementing such IDS’s various benefits are models.
achieved such as in terms of infrastructure, Security,
Scalability, Reliability. II. A DATABASE-CENTRIC ARCHITECTURE
I. INTRODUCTION DAID (Fig 1.) shares many aspects of the AMG
(Adaptive Model Generation) architecture [8]. As in
Intrusion detection is an area growing in relevance AMG, a database component plays a key role in the
asmore and more sensitive data are stored and architecture. Unlike AMG, where the database is only a
processed innetworked systems. The goal of intrusion centralized data repository, in DAID, all major
detectionsystem (IDS) is to provide another layer of operations take place in the database itself
defenceagainst malicious (or unauthorized) uses of
computer systems by sensing a misuse or a breach of a
securitypolicy and alerting operators to an ongoing (or,
at least, recent) attack. Data mining-based IDSs require
lessexpert knowledge yet must provide good
performance. Data mining-based intrusion detection
systems can be classified into misuse detection and
anomaly detection [19]. Misuse detection attempts to
match observedactivity to known intrusion patterns.
This is typically a classification problem. Anomaly
detection attempts to identify behaviour that does not
conform to normal behaviour. This approach has a
better chance of detecting novel attacks. IDSs can also
be distinguished on the basis of the audit data source
(e.g., network-based, host-based).
An effective production-quality IDS need an array
of diverse components and features, including: Fig. 1: Database-centric Architecture for Intrusion Detection
• Centralized view of data DAID also explicitly address data
• Data transformation capabilities transformations,an essential component in analytics.
• Data mining methods DAID has the following major components:
• Real-time detection and alert infrastructure • Sensors
• Reporting capabilities • Extraction, transformation and load (ETL)
• Data analysis • Centralized data warehousing
• System availability • Automated rule generation
This paper demonstrates that the Oracle Database, • Real-time and offline detection
with its capabilities for supporting mission critical • Report and analysis
applications, distributed processing, and integration of The activity in a computer network is monitored by
analytics, can be an appropriate platform for an an array of sensors producing a stream of audit data.
IDSimplementation. Given the data-centric nature of
The audit data are processed and loaded in a evaluationprogram (IDEP)” in Lincoln Laboratory.
centralized data repository (ETL). The stored data are Since 1999, KDD’99 was built based on the data
used for rule generation. The rule generation data captured in DARPA’98. This dataset consists of three
mining methods are integrated in the database component and it contains a number of connection
infrastructure–no data movement is required. The records where each connection is a sequence of packets
incoming audit data will be checked by the rules to containing values of 41 features and labelled as either
detect the intrusions. The stored audit data can be also normal or attack.
further investigated using reporting and analysis tools.
The major benefits of using such an integrated TABLE 1: ATTARCK TYPES FOUND IN KDD’99 DARPA DATASET
approach are improved security, speed, data
management and access, and ease of implementation.
Rule generation A number of data mining
techniques have been found useful in the context of
misuse and anomaly detection. Among the most popular
techniques are association rules, clustering, support
vector machines (SVM), and decision trees [1, 15, 17,
10]. In this IDC prototype we have used association rule
mining technique. Association rule mining is one of the
important topics in data mining research. This approach
determines interesting relationships between large set of
data items. This technique was initially applied to the
so-called market basket analysis, which aims at finding
regularities in shopping behaviour of customers of
supermarkets [27]. Agrawal in [28] has proposed the
Apriori algorithm to find quickly Boolean association TABLE 2: NAME OF 41 ATTRIBUTES IN KDD DATESET
rules. In contrast to Boolean association rules, which
handle only simple item-based transactions, the next
generation of association rules faced quantitative
attributes which their values were elements of
continuous domains such as real number domain R.
But, the typical Apriori algorithm was not capable of
dealing directly with such attributes.
A. Apriori Algorithm
As mentioned before, Agrawal proposed the
Apriori algorithm that is known as a fundamental
algorithm for mining frequent item sets in a set of
transactions. He defined the formal statement for
association rules as follows: Let I= { i1, i2, i3,..., im} be
a set of items and D be a set of transactions, where each
transaction T is a set of items such that T I. A set of
items X I is called an Item set. It can be said that T
contains an item set X, if X I. An association rule is an
implication of the form X Y, where X I, Y I and X Y =. IV. CONCLUSION
The rule X Y has Support S (S = support (X Y)) in the
transaction set D, of S% of transactions in D contain X As we are using the database as the platform for the
Y and it has Confidence C (c = support(X implementation of the IDS, it’s easy to generate the data
Y)/support(X)) in the transaction set D, if C% of analysis results and reports. Collected audit data, rules
transactions in D that contain X also contain Y. Apriori generated, finding the intrusions, analysis and reports
algorithm works iteratively and has two steps: In the can be seen using crystal reports and visualization tools
first step, it finds all the large item sets and in the next in.NET.
step, it uses the large item sets to generate effective This allows the lengthy application development
association rules. A more detailed description of these process and provides a standardized and easily
two steps canbe found in [28, 29]. customized report generation. The IDC prototype
leverages the tools available in.NET.
III. INTRUSION DATABASE AND On the rule generation page (Figure 3) admin can
PREPROCESSING OF FEATURES generate the rules by double clicking on the training
In 1998, Defense Advanced Research Project data with a specific type of attacks and use this rules for
Agency (DARPA) funded an “intrusion detection finding the intrusions in the test data.
Abstract—Long Term Evolution (LTE) is the next a single set of Core Network and Radio resources. So,
step beyond the third generation mobile networks based Multimedia Broadcast and Multicast Service (MBMS)
on Third Generation Partnership Project. LTE includes has been standardized as a key feature in Third
enhanced multimedia broadcast and multicast services Generation Partnership Project(3GPP) systems to
(MBMS), also called as Evolved multimedia broadcast and
multicast services (eMBMS) where the same content is
broadcast and multicast multimedia content to multiple
transmitted to multiple users in one specific area. eMBMS mobile subscribers via MBMS radio bearer
is a new function defined in 3GPP Release 8 specification service.MBMS is a point-to-multipoint (PTM) Standard,
that supports the delivery of content and streaming to whose further evolvement and enrichment attracts
group users into LTE mobile networks. In LTE an nowadays widespread interest.
important point of demanding multimedia services is to Long Term Evolution (LTE) has been designed to
improve the robustness against packet losses. In this support only packet-switched services. Long Term
direction, in order to support effective point-to-multipoint Evolution (LTE) provides both the transmission mode
download and streaming delivery, 3GPP has included an single-cell MBMS, MBMS services which are
Application Layer Forward Error Correction (AL-FEC) transmitted in a single cell and multi-cellular evolved
scheme in the standard eMBMS. The standard AL-FEC MBMS transmission mode, providing synchronous
system is based on systematic, fountain Raptor codes. MBMS transmission from multiple cells, also known as
Raptor coding is very useful in case of packet loss during
multicast/ broadcast single frequency network mode of
transmission as it recover the all video frame back from
insufficient data at receiver terminal.In our work, in
transmission. To transmit the same data to multiple
response to the emergence of an enhanced AL-FEC recipients allows network resources to be shared.
scheme, a raptor code has been implemented and MBMS extends the existing architecture with the
performance is evaluated on image and the simulation introduction of 3GPP MBMS bearer service and MBMS
results are obtained. user services. MBMS user services are constructed
above the MBMS bearer service. For the delivery of
Keywords: Long term evolution; multimedia broadcast MBMS-based services, 3GPP defines three functional
multicast services; forward error correction; raptor codes layers. The first layer, called Bearers, provides a
mechanism to data transmission over IP. Bearers based
I. INTRODUCTION on point-to-multipoint data transmission (MBMS
Nowadays there is a significant demand for bearers), which can be used in conjunction with point-
multimedia services over wireless networks due to the to-point transmission. The second layer is called
explosive growth of the multimedia internet delivery method which offers two modes of content
applications and dramatic increase in mobile wireless delivery: download method of discrete objects and the
streaming method providing continuous media. Delivery
access. It is, therefore, foreseen that the wireless
also provides reliability with FEC. The third layer (User
systems will have to support applications with
Service/ Application) Enables applications to the end-
increased complexity and tighter performance user and allows him to activate or deactivate the
requirements, such as real-time video streaming. service.
Furthermore, it is expected that popular content is Generally an MBMS session includes the following
streamed not just to a single user, but to multiple users three phases.
attempting to access the same content at the same
time. This is addressed by standardization bodies A. User Service Discovery Phase
through introduction of a point-to-multipoint service -
MBMS services are advertised to the end user
enhanced Multimedia Broadcast Multicast Service
using 2-way point-to-point TCP-IP-based
(eMBMS), a resource-efficient transmission scheme
communication or 1-way point-to-multipoint UDP-IP-
targeting simultaneous distribution of multimedia
based transmission.
content to many user devices within a serving area, over
complexity even more, we can decrease the reliability compression might be applied. In the Radio Link
of the decoder. Thus, we would have a reduced degree Control (RLC) the resulting PDCP- Protocol Data Units
distribution resulting linear time encoding and decoding (PDUs), generally of arbitrary length, are mapped to
complexity. However, the decoder cannot decode all fixed length RLC-PDUs. The RLC layer operates in
the input symbols with the lower degree distribution for unacknowledged mode as feedback links on the radio
the same overhead constraint. Therefore, utilizing an access network are not available for point-to-multipoint
erasure correcting pre-code would then correct the bearers. The RLC layer is responsible for mapping IP
erasures arising from the weakened decoder. If the pre- packets to RLC SDUs.Functions provided at the RLC
code is a linear time block code, like an LDPC code, layer are for example segmentation and reassembly,
Raptor Codes provide marvelous encoding and concatenation, padding, sequence numbering,
decoding speeds while providing near optimal reordering and out-of-sequence and duplication
performance for the BEC [8]. detection. The Medium Access Control (MAC) layer
Raptor codes are an extension of the other part of maps and multiplexes the RLC PDUs to the transport
LT codes combined with a system of pre-coding. The channel and selects the transport format depending on
design and degree distribution pre-coding is the heart the instantaneous source rate. The MAC layer and
of Raptor codes. Instead, the media data is protected physical layer appropriately adapt the RLC-PDU to the
using the application layer FEC with Raptor codes. expected transmission conditions by applying, among
To systematically increase the reliability of the others; channel coding, power and resource assignment,
transmission, an application layer FEC code can be and modulation. [1]
used.
The rest of this paper is organized as follows: in 2) MBMS download (file) delivery protocols stack
Section II we provide an overview of the 3GPP AL- MBMS download delivery method aims to
FEC eMBMS delivery framework and Section III distribute discrete objects (e.g. files) by means of a
presents a detailed description of the examined AL- MBMS download session. Download method uses the
FEC scheme. In Section IV we present the simulation File deLivery over Unidirectional Transport (FLUTE)
environment and the conducted experimental results. protocol when delivering content over MBMS bearers.
Finally, in Section V we draw our conclusions and we FLUTE is built on top of the Asynchronous Layered
describe some possible future steps. Coding (ALC) protocol instantiation. ALC combines
the Layered Coding Transport (LCT) building block
II. EMBMS PROTOCOL STACK and the FEC building block to provide reliable
asynchronous delivery of content to an unlimited
A. 3GPP AL-FEC eMBMS Delivery
number of concurrent receivers from a single sender. A
The 3GPP standard multicast services, MBMS detailed description of the FLUTE building block
mentioned [1], is a service of point-to-multipoint way structure can be found in [1]. Thereafter, FLUTE is
in which data is transmitted from a single source to a carried over UDP/IP, and is independent of the IP
group of several mobile terminals in an area of version and is forwarded to the Packet Data
specific service. 3GPP defines two delivery methods Convergence Protocol (PDCP) layer. The packets are
namely, downloading and streaming. eMBMS user then sent to the Radio Link Control (RLC) layer.
plane stack of these delivery methods is illustrated in The RLC
Fig. 1.
1) MBMS streaming delivery protocols stack
The purpose of the MBMS streaming delivery
method is to deliver continuous multimedia data
(i.e. speech, audio, video) over an MBMS bearer.
MBMS makes use of the most advanced multimedia
codecs such as H.264 for video applications and
enhanced Advanced Audio Coding (AAC) for audio
applications. Real-time transport protocol (RTP) is the
transport protocol for MBMS streaming delivery.
Real- time transport protocol (RTP) is the transport
protocol for MBMS streaming delivery. RTP provides
means for sending real-time or streaming data over user
datagram protocol (UDP), the resulting UDP flows are Fig. 1: eMBMS Protocol Stack
mapped to MBMS IP multicast bearers. Then IP
packets are processed in the Packet Data Convergence Layer functions in unacknowledged mode. The
Protocol (PDCP) layer where for example header RLC layer is responsible for mapping IP packets to
RLC SDUs. The Media Access Control (MAC) Layer needed to decode the source symbols is exactly the
adds a 16 bit header to form a PDU, which is then sent number of source symbols, the Raptor code has a
in a transport block on the physical layer. performance close to that property. The performance of
an AL-FEC code can be described by the decoding
III. AL-FEC SCHEME failure probability of the code. The study presented in
[12] describes the decoding failure probability of
In this section we provide a detailed description
Raptor code as a function of the source block size and
of the 3GPP standardized Raptor code.
the received symbols. In fact, the inefficiency of the
A. Raptor Codes Raptor code can accurately be modeled by (1) [12]
IV
V. RAPTOR CODE
O IMPLEMEN
NTATION
Fig. 6: The Recovered
R Image aafter Raptor Deco
oding
The performance of Raptor codes has been
evaluated as shown inn figure 3. The encoderr and
V. CON
NCLUSION & FU
UTURE WORK
K
decoder arre designed as specified in section IIII. A
program off random packket loss has allso been develloped In this work, w we have evalu uated the
to analyze the coding Raaptor yield. perfoormance of AL-FEC
A baseed on the Raaptor codes
provvided by IET TF RFC 50533 to providee reliability
againnst packet Loosses in Imaage. Thus, Raptor coding
signiificantly im mproves the image quality q by
recovvering the losst packets falliing on the tran
nsmission.
Some possiblle future actioons that may result from
this work may bee possible connsideration iss the design
and evaluation of an adapptive algoritthm which
calcuulates the opptimum FEC encoding pro ocess. This
mech hanism couldd be based onn a system of o feedback
reports on netwoork conditionns, as proposed in the
Fig. 3: Block Diagram
D of Raptorr Coding databbase and repoorts coding paarameters AL-FEC could
We have
h used im
mage as a soource where each be adjusted. Fiinally, recen tly appeared d AL-FEC
pixel of the image is considereed as a sym mbol. schemme, named RaptorQ
R couuld further enhance
e the
Simulationn Parameters are Symbolss K=1600, Each searcch field of relliable multicasst in mobile networks.
n
196 Bhag
gwant Institute of
o Technology, Muzaffarnagar,
M Uttar Pradesh, India
AL-FEC Raptor Code Implementation Over 3GPP eMBMS Network
ACKNOWLEDGMENT [7] M. Luby, “LT codes”, Proc. 43rd Ann. IEEE Symp. On
Foundations of Computer Science, Nov. 2002, pp. 271–282.
We would like to thank Yashesh Buch for [8] A. Shokrollahi, “Raptor Codes”, in Proc. IEEE Int. Symp.
providing such an interesting topic. We are also grateful Information Theory, Chicago, IL, Jun. /Jul. 2004, p. 36.
[9] A. Shokrollahi, “Raptor codes,” Information Theory, IEEE
to Santhi Chinnam for valuable discussions and friendly Transactions on, vol. 52, no. 6, pp. 2551–2567, June 2006.
support. [10] M. Luby, A. Shokrollahi, M. Watson, and T. Stockhammer,
“Raptor forward error correction scheme for object delivery,”
REFERENCES September 2007, Internet Engineering Task Force, RFC 5053.
Available at http://tools.ietf.org/html/rfc5053.
[1] 3GPP TS 26.346 V10.4.0.,” Technical Specification Group [11] Luby M, Watson M, Gasiba T, Stockhammer T, Xu W.”
Services and System Aspects: MBMS, Protocols and codecs”, Raptor codes for reliable download delivery in wireless
(Release 10), 2012. broadcast systems.” In Proc. of CCNC, 2006.
[2] P. Elias. Coding for two noisy channels. in Proc. 3rd London [12] T. Stockhammer, A. Shokrollahi, M. Watson, M. Luby, and T.
Symposium on Information Theory, pages 61–76, 1955. Gasiba,“Application Layer Forward Error Correction for mobile
[3] Han Wang, “Hardware Design for LT Coding”, MSc Thesis, multimedia broadcasting,” Handbook of Mobile Broadcasting:
University of Delft, Holland, 2006. DVB-H, DMB, ISDBT and Media Flo, CRC Press, pp. 239–280,
[4] Berlekamp, E.R.: “Algebraic coding theory”, McGraw-Hill, 2008.
New York, 1968. [13] Erol Hepsaydir,Emmanuel Witvoet,Nicola Binucci, Sachin
[5] M. Luby, M. Mitzenmacher, A. Shokrollahi, and D. Spielman, Jadhav, “Enchanced MBMS in UMTS Networks and Raptor
“Efficient erasure correcting codes”, IEEE Trans. Inf. Theory, Codes”, The 18th Annual IEEE International Symposium on
vol. 47, no. 2, pp. 569–584, Feb. 2001 Personal, Indoor and Mobile Radio Communications
[6] David J. C. MacKay, “Fountain Codes”, Communications, IEEE (PIMRC'07)
proceedings vol. 152, no. 6, Dec. 2005, pp. 1062 – 1068.
Abstract—An ad-hoc network is one which is joined There may be used many nodes to transfer a
by multi-hop wireless links called nodes they are allow to message from sender to receiver and each node treat as
move that’s why such type of network topology is known router and host at the same time. Where the
as dynamic topology. In this topology, there is no communication between the transmitting and receiving
centralized infrastructure support. We can use both
reactive and proactive routing techniques to find the
node is take place by a special node, i.e., called access
routes. point and it can be also pointed out the node which is
In case of proactive routing technique, ad hoc used to find out the path of the network is called routing
network maintain a routing table with routes to all other node or router (switches). Because its departure and
nodes in the network. Routing table will occupy a large new arrival nodes of the routes change time to time
amount of space if the network is large and fast changing that’s why it is called dynamic network and there is one
of node position lead to in efficient usage of network another challenge is to face to maintain up to date
resources however an advantage there, i.e., routes between routing information of this dynamic network. For this, it
arbitrary source destination pair are readily available. can be done by choosing any one of these: reactive and
While Some algorithm like reactive techniques, suffer
from high latency hence more time consume to find the
proactive routing technique.
route that search a route on demand usually by flooding In the mobile ad-hoc network, routing protocols
the network with route request packets and excessive (proactive and reactive) have to face many problems
flooding can lead to network clogging (Blocking). such as fast changing node position, low transmission
Besides reactive and proactive protocol MDVZRP power and asymmetric links. Here the routing protocols
protocol can be used to reduce the latency. MDVZRP is a prove it inefficient under these circumstances. To
hybrid protocol which is the extension of MDSDV overcome these problems we can use a hybrid multipath
(Multipath destination sequenced distance vector). In routing protocol ‘MDVZRP’. It extends from the
MDVZRP, all routes in the routing table are active and MDSDV (multipath destination sequenced distance
useable. There is no need of periodic updates of routing
table or no need to find the route unless a broken link has
vector).
been discovered or a new node come into existence. In proactive protocols (table-driven) uses the
In this paper we are going to compare the bandwidth to maintain routes information [1], if we
performance of MDVZRP, AODV and DSDV at different want some change in the network topology throughout
parameter like CBR. These protocols can be simulated on the whole network on the other hand the reactive
ns2 platform. protocols involves the long request delay and it is also
prove inefficient for route determination in the entire
I. INTRODUCTION network. To overcome these problems ZRP extends a
hybrid protocol combines the best properties of both
We are using the wireless cellular systems since the
proactive and reactive protocols but it is also access the
1980. After that we have seen the first, second and third bandwidth in the same manner which is used in
generation of wireless communication systems. Now a proactive and reactive protocols. In MDVZRP, we
day’s every person use wireless communication system assume that all the nodes or routes are active, usable
directly or indirectly. Majority of the message transfer and we can update it at only need when the message
are based on the point to point communication however packet is not able to reach at the destination due to the
some of them are based on multipoint communication. broken link or non reachable link or a node received a
In last decade wireless communication achieve great new routing information (new route), the source receive
achievement in the field of ad-hoc network. In ad hoc a message to indicate it. Therefore, MDVZRP reduce
network mobile nodes are present between source and the proactive and reactive scope. It shows reactive
destination to maximize the utility of communication nature whenever an unknown route is to be found out.
The main goal of MDVZRP is to select the optimum
network. In other words we can say mobile ad hoc
available path when it runs some filtering procedure at
network is an autonomous network for mobile nodes entries. In it there is no need for periodic updates of the
which are connected via wireless links. routing table, hence control traffic also reduced.
MDVZRP uses a topological map of the zone centred hybrid protocol that utilizes the best property of both
on a node. In the nodes route are immediately available reactive [2][3] and proactive approach. In MDVZRP
while outside the zone the route is to be found out. protocol all the routes are in active mode and can be
Unfortunately, up to now a full simulation tool for utilize whenever required. When a new node join the
the next generation networks has not been realised,So network, it send an update message and on the other
our research efforts have been focused on the hand when a node find a broken link, send an error
implementation of software modules for the network message hence partial updating is required only.
simulator (NS2) [6] Therefore it reduces the scope of proactive protocols. In
MDVZRP, a node has a flat view over the whole
II. NETWORK SIMULATOR network. Whenever a node joins to network that
completes its routing table from its nearest neighbour.
The three routing protocols AODV, DSDV and
In case of receiving a broken link message (error
MDVZRP are used in the radio propagation and the message) node does not send a route request every time
simulation model of these protocols can be made by while it utilize another optimum path stored in routing
using NS2.
table and saves time and bandwidth.
ABOUT NS2 NS is an object oriented simulator.
NS use two languages because the simulator works in
IV. COMPARISION OF MDVZRP TO PROACTIVE,
two different ways when needed. The two languages are
REACTIVE AND HYBRID ROUTING PROTOCOL
C++ and Otcl (object tool command language) which
are used in the NS. C++ is fast to run but it cannot DSDV (Destination sequenced distance vector)
change immediately. It is suitable for detailed protocol require a periodic updates that causing an excess
implementation. Otcl works opposite to the C++ flooding of route request messages. On the hand, In
languages because Otcl is slower to C++ but it can be MDVZRP, each node keeps all optimum paths in their
changed quickly and it is used for simulation zone radius without periodically updating the routing
configuration. table. Hence we can expect that MDVZRP will have
The otcl is also used to write the fronted program significantly fewer update messages then proactive
of NS2. While the backend of NS2 simulator is written protocols.
in the C++ language. CBR (constant bit rate) file and
scenario file also define the traffic and movement
pattern in separate files. A tcl program can be create in
the CBR file called cbrgen.tcl which is present in the
directory “ns2 /inded-utils/cmu-scen-gen/”. The setdest
present file is an.EXE file which is used to define the
movement in the folder “Ns2/ inded-utils/cmu-scen-
gen/setdest”.
To define the traffic and movement pattern in
separate files which named are CBR file and scenario
file. These files are generated by using the following
commands in the appropriate directory respectively
In this scenario, the traffic engineering (TE)
introduce about the multiprotocol label switching Fig. (a): An Example of 10 Node Network
(MPLS), but it is not able to give any QOS guarantee to TABLE 1: THE WAY TO SELECT SHORTEST PATH
MOIPS (multimedia application over IP). Traffic FROM SOURCE TO DESTINATION
engineering capabilities can be exploit new information
S.N. Route Node Number of Remark
as link bandwidth availability which is provided is Sequence Hoping
extensions of traditional routing protocols 1 1-2-4-3-8-9 5
Ex- OSPF, IS-IS and RSVP or (CR-LOP) like 2 1-3-8-9 3 Shortest path
3 1-4-7-8-9 4
constraint–based label distribution protocol traffic
4 1-2-4-7-8-9 5
engineering extensions (RSVP-T), allow the network to 5 1-2-4-3-7-8-9 6
provide a differentiated treatment in terms of QOS to 6 1-2-4-7-3-8-9 6
different connections. 7 1-2-5-4-3-8-9 6
8 1-2-5-4-7-8-9 6
9 1-2-5-4-7-3-8-9 7
III. MDVZRP MOTIVATION 10 1-2-5-4-3-7-8-9 7
11 1-3-4-7-8-9 5
Proactive protocol require extra bandwidth to
12 1-3-7-8-9 4
maintain the route map however Reactive protocol 13 1-4-7-3-8-9 5
involves long route request delays. MDVZRP is a 14 1-4-3-7-8-9 5
In reactive protocol like AODV [4], a route TABLE 3: ROUTES ARE OBTAINED FROM THE UPDATE
MESSAGE SENT BY NODE 4
discovery process is done at every time whenever a
specific destination is given. Each node put a signature Node 2:
on the route request reply message and node 1st Hop 2nd Hop Metric Destination
4 6 2 6
select a route which has lowest signature, means lowest
Node 3:
distance path. 1st Hop 2nd Hop Metric Destination
4 6 2 6
V. ROUTING INITIALIZATION Node 5:
1st Hop 2nd Hop Metric Destination
Whenever, a new node come into the existence in 6 - 1 6
the range of network, route initialization process is 4 6 2 6
going on. In this process, a beacon message or
An update message was sent by the node 5 to the
sometimes called a Hello message is sent to its nearest
1st hop neighbour is only node 4. Table (4) shows the
node to connect with network. This new node is
new entry that node 4 has obtained from node’s 5
completely dumped in initial state because it does not
update message.
have its own routing table. In next fig (b), we assume
that node 6 is a new node and all the nodes are TABLE 4: ROUTES ARE OBTAINED FROM THE UPDATE MESSAGE SENT
BY NODE 5
symmetrical means they have same range to capture
another neighbour node. Node 4:
1st Hop 2nd Hop Metric Destination
6 - 1 6
5 6 2 6
TABLE 13: ROUTING TABLE OF NODE 4 AFTER RECEIVING In the future work, we will evaluate and compare
THE RERR MESSAGE
MDVZRP to DSDV, AODV and ZRP. There are the
1st Hop 2nd Hop Metric Destination some metrics on the basis of these we can evaluate
2 1 2 1 MDSDV and we can simulate these on Ns2. where we
3 1 2 1 expect it will achieve:
2 - 1 2
3 - 1 3
1. Low overhead than the standard protocols:
4 - 0 4 • Latency is less than AODV by (N(c/b+t)-
5 - 1 5 tR).
6 5 infinite 5 Where:
5 6 infinite 6 N= number of nodes from hop
6 6 1 6
c/b= control packet size
TABLE 14: ROUTING TABLE OF NODE 6 AFTER RECEIVING t= route hunting time
THE RERR MESSAGE
R= radius
1st Hop 2nd Hop Metric Destination • Routing table size less than DSDV.
4 2 3 1 • Less control traffic than ZRP.
4 2 2 2
2. Network performance vs radius size regarding
4 3 2 3
4 - 1 4 to evaluation metrics data throughput, packet
5 4 2 4 delivery ratio, routing overhead and average
5 - 1 5 packet delay.
4 5 2 5
6 - 0 6
REFERENCES
VIII. CONCLUSIONS AND FUTURE WORK [1] Perkins, C.E., Bhagwat, p., “highly dynamic destination
sequenced distance vector routing (DSDV) for mobile
In this paper, we studied a protocol named computers”, October 1994, computer communication pp. 234-
244.
MDVZRP (Multipath Distance Vector Zone Routing [2] Nick las Beijar: zone routing protocol (ZRP), networking
Protocol). MDVZRP is a hybrid routing protocol which laboratory, Helsinki university of technology P.O. box 3000,
extends proposed protocol Multipath Destination FIN -02015 HUT, Finland, Nicklas. Beijar @ hut.fi
Sequenced Distance Vector (MDSDV). This protocol [3] Haas, Zygmunt J., Pearlman, Marc R,: a new routing protocol
for the reconfigurable wireless networks, proceeding of 6th IEEE
can be broadly classified into three categories, namely, international conference on universal personal communication,
proactive, reactive and hybrid. There are the two IEEEICUPC ’97.San Diego, California, USA October 12 16
techniques to send the packets over symmetrical links- 1997
• Broadcast Technique [4] Peter J.B King, A. Etorban and Idris skloul Ibrahim: Multipath
Destination Sequenced Distance Vector (MDSDV), the 8th PG
• Unicast Technique Net, Liverpool john Moores University, UK 28-29 June 2007,
With the help of Hello message and Full dump, pp. 93-98
nodes create their routing tables for saving multi [5] Idris Skloul Ibrahim, A Etorban, Dr Peter J.B King School of
Mathematical and Computer Sciences (MACS) Heriot Watt
optimum paths. University at Edinburgh UK.
If a node wants to communicate with other node [6] The Network Simulator Version 2 (NS2) Home
but it has no route at this time MDVZRP will try to Page:http//www.isi.edu/nsnam/dist/
provide such a route. It provides an alternating
path when the primary path breaks. It also gives the
node ability to get information from any route pass
through it.
Abstract—Botnet is one of the malicious threats A centralized botnet has a central server. IRC
known to academic community. It provides serious threat protocol was originally designed for large area to
to cyber security. Behavior of the threat has never been allow for multiple forms of communication and data
methodically studied. It uses command and control scattered among large number of end-machines. The
channels which operate over variety of network topologies
and communication mechanism to update and direct
traditional common control infrastructures were mostly
themselves.This paper is a survey of botnet size based on IRCs (Internet Relay Chat). A botmaster
measurement techniques. The survey clarifies botnet establishes an IRC server and gives out commands
phenomenon and discusses botnet measurement techniques using a special channel. Another C&C based botnet
with a brief comparison of techniques. The survey also used depends on HTTP server for distributing
classifies botnet measurement techniques into two classes: commands. In this case, the bots periodically contact
Active measurement and Passive measurement. server to obtain new commands and act upon them.
Keywords: Bot; Botnet; Botnet measurement techniques Both IRC-based and HTTP-based C&C provides a
centralized mechanism [8].In a centralized C&C
I. INTRODUCTION infrastructure; there are usually command-and-control
servers, under the control of the botmaster. Because all
A compromised computer secretly controlled is bots connect to these servers, botmasters are able to
known as a bot and malicious people controlling the communicate with the bots simultaneously and can
compromised computer are known as botmaster. Like give out commands to all the bots that are connected to
the previous generations of viruses and worms, a bot is the botnet [15]. These may be used for direct feedback;
a propagating application that infects vulnerable hosts easy monitoring of the botnet status for the botmaster
through illegal activities to expand for harmful purpose and giving information about fundamental properties
[8]. Another definition defines bot as a malicious like number of active bots. IRC-based botnets are
programe installed on a victim machine; Collection of intrinsically flexible and scalable [15].
these bot forms a network called botnet [2], commanded
to carry out attacks. There is a Control channel required
to send commands to bot obtaining result and status
messages. Through command and control (C&C)
botmaster controls bots. Botnet can be classified as IRC-
based, HTTP-based, DNS based or Peer to Peer (P2P)
botnets according to their C&C architecture [12]. Fig. 1
shows typical botnet architecture [2].
advantage of this technology being botnet architecture bot or may be for some changes in the bot
which constitutes only control channel and no C&C army. Moreover, sometimes the updated
server, thus significantly improving the result. Fig. 3 binary moves bots to a different C&C server.
shows peer to peer botnet. This process is called server migration and it
is very useful for botmasters to keep their
botnet alive. Botmasters try to keep their
botnets invisible and portable by using
Dynamic DNS (DDNS) which is a resolution
service that facilitates frequent updates and
changes in server locations. In case authorities
interrupt a C&C server at a certain IP address,
the botmaster can easily set up another C&C
server instance with the same name at a
different IP address. IP address changes in
C&C servers propagate almost immediately to
bots using short time-to-live (TTL) values for
the domain names setup by DDNS providers.
Consequently, bots migrates to the new C&C
server location and stays alive.
Fig. 3: Peer-to-Peer Botnet [15]
II. BACKGROUND
A botnet can be understood by its life-cycle of
botnet [12], as how a botnet is created and infects
systems across infrastructure. This life-cycle is depicted
in Fig. 4.
• Infection phase: In the initial infection phase,
the attacker, scans a target subnet for a known
vulnerability, and infects victim machines
through different exploitation methods.
• Secondary phase: After initial infection, the
infected hosts’ written script, known as shell-
code, is executed in secondary phase. The
shell-code fetches the image of the actual bot
binary from the specific location via FTP,
HTTP, or P2P. The bot binary installs itself on
the target machine. After the installation of bot
program, the victim computer becomes a
Fig. 4: Botnet Life Cycle [12]
“Zombie” and runs the malicious code. Each
time the zombie is rebooted, bot application
III. THREAT OF BOTNET
starts automatically.
• Connection phase: In connection phase, the bot Some attacks of botnets are defined. A major threat
program establishes a command and control is obtaining a profit by stealing user data and password
(C&C) channel and connects the zombie to the from infected hosts. Key targets include passwords for
C&C server. various services like e-mail accounts, web shops,
• Command & control phase: After connection banking platforms or social networking platforms. This
phase, the actual botnet C&C activities are process is often called identity theft. For example, an
started. Zombie is a part of botnet during the e-mail login can reveal the owner‘s contacts through
setup of control channel. The botmaster uses inbox and address books.
the C&C channel to give out commands to his Other more popular use of botnet is for
bot army. Botmaster sends bot programs for unsolicited mailing, also known as spamming [5, 15].
receiving and executing commands. The C&C While classic spamming has been achieved with single
channel is used for sending the command and computers or comparatively small networks owned
bots are commanded to download the updated directly by the spammers, botnets have enabled them to
binary. perform a cost shift, in terms of computation power,
• Maintenance & update: Bot controller may bandwidth and reputation, towards the owners of
update the bot binary to hide the detection of compromised computers. Also, the ability of botnets to
use bot’s IP address to hide the true originator of the highly scalable. Additionally, special measures must be
spam email complicates countermeasures used such as implemented to prevent any part of the system from
blacklisting of suspicious IP addresses. The primary participating in malfeasance [1].
intention behind mailing is advertising. Commonly,
third parties buy a batch of spam mails for delivery IV. MOTIVATION
from botmasters in order to advertise their products. In
addition, spam mails can be used either as a malware Botnets have become a serious threat, now-a-days.
vector with attachments or as links to malicious A clear understanding of the network is required to
understand botnet. We focused on counting botnet
websites. Password fishing, known as phishing, is
another use of spam, in phishing intruder acts as an populations for a number of reasons:
authorized person. • To understand the emerging threats such as
In Click fraud [15], the attacker sets up an account storm, nugache bot, etc.
with an online advertiser, who pays for page visits or • We wanted to understand the scale of the
for additional advertising links, for example, clicking resources we need to gather as we tackle a
on a certain banner. Second, the attacker uses the botnet.
controlled bots to visit those pages and generating • Continuous measurement is vital in order to
clicks on target banners. This is possible because he understand what mechanisms are effective at
has full control of the victim‘s machine and may use reducing the botnet’s population. If the
it to simulate behavior. In this case, the attacker gains number ever drops to zero, we can call it a
money directly from the advertising company. victory.
Botnets usually consist of large number of remote • The main power of a botnet is in its
machines which enable botmasters to start attacks numbers: the bigger it is, the more it can do
against websites. By commanding bots to contact a because of the compounded bandwidth and
website frequently, the servers are rendered computing power of its members.
unreachable as they cannot handle the incoming traffic.
This attack is called a Distributed Denial of Service V. STATE OF THE ART
(DDoS) attack [15], as a large number of Rajeb et al. [1] attempted to clear the fog
geographically- distributed bots are involved in the surrounding botnets by constructing a multifaceted and
attack. These attacks happen regularly and the profit distributed measurement infrastructure. Their data
comes from the fact that many companies depend on collection methodology encompasses three logically
web-based services, e.g. web shops, gambling websites. distinct phases: (1) malware collection, (2) binary
Botnet size is one of the most important analysis via gray-box testing and, (3) longitudinal
characteristics to evaluate the threat of botnet [17]. In tracking of IRC botnets through IRC and DNS trackers.
general, botnet measurement is the process of botnet Throughout a period of more than three months, they
estimation which counts number of bots. They are used this infrastructure to track 192 unique IRC botnets
botnet Footprints and Live Population. Botnet footprint of size ranging from a few hundred to several thousand
refers to the overall size of infected population of infected end- hosts. Results show that botnets represent
botnet at any time in its lifetime. Botnet live population a major contributor to unwanted Internet traffic—27%
is the number of live bots simultaneously present in C2 of all malicious connection attempts observed from
channel [3, 17]. Hence measurement is the process distributed darknet can be directly attributed to botnet
which uses some technique to measure the size of related spreading activity. Furthermore, they
botnet. Botnet size measurement is very important to discovered evidence of botnet infections in 11% of
detect and quantify the effect of the botnet oriented the 800,000 DNS domains examined indicating a high
attacks launched or the number of spams sent by the diversity among botnet victims. They said that a
botnet, the coverage area and region botnet has botnet’s footprint can become fairly large in size
(e.g., more than 15,000 bots).
activated [5]. Methodology of measurement
Zhuge et al. [2] showed with the help of a
encompasses three logically distinct phases: (1)
distributed and fully-automated botnet measurement
malware collection, (2) binary analysis via gray-box
system, able to discover and track 3,290 botnets during
testing and, (3) longitudinal tracking of IRC botnets
a period of almost twelve months. Based on the analysis
through IRC and DNS trackers. The goal of the
of the collected information, gave a broad overview of
malware collection phase is simply to collect as many
the current botnet phenomenon and presented several
bot binaries as possible. However, developing a
characteristics and features of IRC-based botnets
scalable and robust infrastructure to achieve this goal is
observed in the wild. This information can be used to
a challenging problem in its own right, and has been the
get a deeper understanding of the threat posed by these
subject of numerous research initiatives. In particular,
networks. Furthermore, these results can also leverage
any malware in collection infrastructure must support a
research in the area of botnet mitigation.
wide array of data collection endpoints and should be
Rajeb et al. [3] provides a comprehensive list of table (DHT) routing protocol, for the P2P networking.
botnet size estimation techniques and highlights the A bot measuring criteria by analyzing the relationship
challenges associated with each. They denote a botnet’s between the two aliasing, by which we can effectively
footprint as the overall size of the infected population identify Storm bots. They developed an active crawler,
at any point in its lifetime. While this measure reflects Glovnet, to crawl the whole Overnet network. Due to
how wide spread a botnet infection is Second, they the flaws in the Node-ID generation algorithm of the
consider the botnet’s live population as the number of Storm bots, we points out Storm bots partly leads to
live bots simultaneously present in the command and the IP address aliasing and Node-ID aliasing
control channel. Therefore, unlike its footprint, the live phenomena. After analyzing the relationship between
population of a particular botnet indicates the botnet’s these two phenomena, they deduce a Storm bots
capacity to perform a malicious task posted as a measuring. Kang et al. [8] applied enumeration for
command message by the botmaster. This includes the measurement in the peer-to-peer nature of the
techniques that directly count bots connecting to a Storm botnet requires each node to communicate with
particular server. Two main techniques used are other nodes in order to search or publish information,
infiltration and redirection. using multi-hop P2P routing protocol. There are
Holz et al. [4] showed that measurement strategy different ways in trying to enumerate the number of
can be used as a way to disable the communication P2P bots in the Storm network.
within the Storm botnet to a large extent. As a side Jun et al. [9] developed new techniques to measure
effect, able to estimate the size of the Storm botnet, botnet size by using traces of emails from HUST
they were first to introduce an active measurement Mail Server. In this trace, they have successfully
technique to actually enumerate the number of infected identified the size of botnets through URL and
machines, crawl the P2P network, keep track of all collaborative mail servers. They estimate the live
peers, and distinguish an infected peer from a population of the botnet, including IRC botnet and P2P
regular one based on characteristic behavior of the bots. botnet. A new method is proposed to measure the size
Castle et al. [5] present a novel technique for the of the botnet. Email dataset collected from HUST Mail
automatic identification, analysis and measurement of Server is adopted for their analysis. They still cannot
botnets used to deliver malicious email. This technique fully determine the size of the botnet because many
has been very successful with the amount of spam home computers are connected to the internet through
increasing by a significant percentage. In April 2008, ADSL, cable or other devices which make the user’s IP
their research system, which processes around 50,000 changes daily or even hourly Zhuang has proposed an
legitimate emails per month, a figure which has been effective method to estimate IP dynamics. After IP
constant over the last two years, recently reached a dynamic treatment, they can count the number of
level of 97% spam. This contrasts significantly with distinct machines, and then obtain the size of the
April 2007, when the percentage of spam was at 85% botnet.
and in April 2006 where the spam percentage was Binbin et al. [10] considering that the C&C flows
46%.A reference implementation of a system developed related to a P2P-based bot exhibit stability on
to demonstrate these techniques. This system has been statistical meaning due to the impartial position in
deployed in a live environment, and has shown to be botnet and performing pre- programmed control
highly effective in use, provides a technique that uses activities automatically, a novel detection approach
purely passive analysis, requires relatively low amounts based on the control flow stability is proposed. The
of resource, The Plato Algorithm This algorithm measurement of control flow stability is firstly
provides a representation of the template used to create derived from the P2P-based C&C case study and the
the emails. definition of control flow stability.
Binbin et al. [6] proposed a novel method for Plohmann et al. [15] described various botnet
identifying Strom bots using an active crawler that measurement and detection approaches.
can collect location information from all participants. Xin Hu et al. [16] examined measuring and
Due to the flaws in the generation algorithm of the analyzing global IP usage Patterns of domains, this
global identifier of the Storm bots, the location paper considers the global IP-usage patterns exhibited
information of each Storm bot partly contributes to by different types of malicious domains, with a focus
the aliasing phenomena. After analyzing the on single and double fast-flux domains. They have
properties of aliasing phenomena, they deduce a Storm developed and deployed a lightweight DNS probing
bot measuring criteria to identify bots. The results show engine, called DIGGER, on 240 Planet Lab nodes
that the size of Storm Worm botnet would estimate over spanning 4 continents, Collecting DNS data for over
400,000 during the first two weeks of February 2008 3.5 months enabled to measure the global IP usage.
with the counting of online bots between 16,500 and Shangdong et al. [17] investigates the nature of
23,000. The Storm Worm botnet utilizes the Overnet botnet size, upon which four issues are introduced. The
protocol. Which is a Kademlia-based distributed hash paper summarizes the existing solutions to the four
issues and analyzes the challenges to resolve them limitations. First, botmasters may suppress bot
deeply. (1), the measurement of botnet live population, identities from being transmitted to the channel and in
which is a problem of botnet detection in nature. (2), doing so render this technique useless. Second, even
the measurement of botnet footprints. (3), dynamic when this information is available, counting can lead to
tracing of botnet size. (4), area issue of botnet size. For: different estimates depending on whether we count the
(1), live population can be obtained by network fully qualified unique user IDs or the IP addresses—be
anomaly detections, but footprint contains offline bots it cloaked or plain. Temporal population variations due
which cannot be detected by network anomaly to bot cloning and temporary migration of bots
detections. They describe that local size is used to complicate this issue even further. What this means is
estimate first then this size is used to estimation of that it is difficult to provide an accurate bot count in
global size. these cases, as distinguishing between actual bots and
temporary clones or migrants is nontrivial [3].
VI. TECHNIQUES
A. DNS Redirection
In active measuring area, most researchers use
honeynet to infiltrate the botnet getting some As an alternative to botnet infiltration, explored a
information of peers and state of the network, by technique for counting infected bots by manipulating
running an instance of the botnet; and developing a the DNS entry associated with a botnet’s IRC server
crawler using specific protocol queries, they can collect and redirecting connections to a local sinkhole [9]. The
some information, such as node IDs, IPs, and port sinkhole completed the three-way TCP handshake with
numbers. The advantage of this method is that it can bots attempting to connect to the (redirected) IRC
directly get the topology of the network and has a more server and recorded the IP addresses of those victims.
correct result, but the crawler needs a priori knowledge Their results suggest the existence of large botnets
of botnet such as captured bot binaries and the protocol with populations up to 350,000 bots. Unfortunately,
the botnet using, so it will be difficult to operate on a although this approach allows us to observe the IP
new botnet, especially on the encrypted botnet [4, 8, 9]. addresses of different bots, it has a number of
In passive measurement area, the monitors are limitations. First, this technique can only measure the
fixed on the edge of the backbone or the core routers, botnet’s footprint. The reason is that although the
or the bound of the ISP; the monitors can get node sinkhole observes bot connection attempts, it is
numbers, flow characters etc. Using these overview impossible to know how many live bots are
flow information suitable methods can be adapted to simultaneously connected to the actual server channel.
measure botnet. The advantage of this method is that it Second, as the sinkhole does not host an actual IRC
does not require any prior knowledge, but has a high server, there is no way of knowing if the bots are
false positive rate [1, 2, 3, 5]. connecting to the same C&C channel. Finally, it is
The first category includes techniques that directly conceivable that botmasters can detect DNS
count bots connecting to a particular server. Two main redirection and subsequently redirect their bots to
variants are relevant here: infiltration and redirection another IRC server thus distorting the estimate provided
[4]. An obvious way to learn several aspects of a by this technique.
botnet’s activity is to infiltrate the botnet by joining the B. DNS Cache Snooping
C&C channel. Botnet infiltration provides valuable
information about several malicious activities such as The measurement technique called DNS Cache
DDoS attacks. Botnet infiltration technique is used to Snooping is based on the caching property
provide in-depth analysis of several facets of botnets, implemented and used by many DNS servers. If a DNS
including inferring their membership by directly server is queried for a domain for which it has no
counting the bots observed on individual C&C entry defined, it will issue a query towards the
channels. The tracker intelligently mimics the behavior responsible authoritative name server on behalf of the
of actual bots and joins a number of botnets, all the querying client and store the resulting data record
while recording any information observed on the afterwards in a local cache. It is still possible to
command and control channel. This information may estimate a botnet’s size by exploiting external
include the identities of all active bots. In this case, the information. In this case, however, techniques that rely
botnet’s footprint is simply the total number of unique on externally visible information can only provide an
identities observed on the channel over the entire estimate of a botnet’s footprint [15]. A natural source
tracking period. Similarly, the botnet’s live population of externally visible information about a botnet’s
is measured by counting the number of bots prevalence is DNS. Use of DNS cache snooping is to
simultaneously present on the channel at a particular uncover a botnet’s footprint. In short, this approach
time. In some cases, this estimate can also be derived takes advantage of the fact that bots normally make a
from the IRC server’s welcome message. Despite its DNS query to resolve the IP address of their IRC server
simplicity, this technique suffers from a number of before joining the command and control channel. Our
technique estimates a botnet’s DNS footprint by These observations raise an important question:
probing the caches of a large collection of DNS servers While counting botnet members, are we really counting
and recording all cache hits. A cache hit implies that at actual compromised machines? Although direct
least one bot has queried its name server within the time counting of bots by botnet infiltration seems to be the
to live (TTL) interval of the DNS entry corresponding most direct way of estimating a botnet size, it is unclear
to the botnet server. The total number of cache hits whether or not the resulting estimate is a count of real
provides an indication of the botnet’s DNS footprint. bots. For one, temporary botnet migrations can
That said, a botnet’s DNS footprint provides only a significantly inflate the membership of a particular
lower bound of its actual footprint. For one, using DNS botnet. If counting the population of Botnet,
to estimate size is only possible for DNS servers that immediately after the migration arrives at an inflated
allow probes from arbitrary clients and reply to count. While on the surface this may not seem as a big
queries for cached results. To yield an accurate concern, if such migrations occur frequently, then that
estimate, this technique requires a large list of such could be substantially over counting the cumulative bot
servers. Moreover, botnet servers that have DNS names population.
with low TTL further complicates this technique For further description about the impact of bot
because, for such names, the probability of a cache hit cloning on size estimation, clone commands observed
from an infected domain is low. Finally, a hit indicates in IRC traces of the botnets are also tracked.
only the existence of at least one infected host
associated with that DNS server. Recently, suggested VIII. CONCLUSION
another DNS-based technique. Their approach infers
bot counts by observing DNS lookups for hosts listed Botnets pose a significant and growing threat
in DNS-based black hole list. The rationale behind this against cyber-security as they provide a key platform
approach is that botmasters tend to query these lists to for many cybercrimes such as Distributed Denial of
detect if their bots are blacklisted and thereby unusable Service (DDoS) attacks against critical targets,
for certain tasks (e.g., sending spam e-mails). This malware dissemination, phishing, and click fraud.
approach has the potential to provide an overall Despite the long presence of malicious botnets, only
estimate of possible bots in DNS-based black hole list, few formal studies have examined the botnet problem
but clearly it cannot estimate the footprint or the live and botnet research is still in its infancy. This paper
population of a specific botnet. surveys botnet size measurement. As mentioned,
highlight values of a botnet is its ability to provide
VII. CHALLENGES anonymity through the use of a multi-tier command and
control (C&C) architecture. Diversity of botnet
The size of a botnet is an important metric to
protocols and structures make botnet measurement a
evaluate the threat posed by it. In general, it is hard to
measure the real size of a botnet since attackers can challenging task. In this survey, botnet measurement
obfuscate this information by modifying the C&C techniques are based on active and passive techniques
server or other means. To obscure the size of the botnet, However, most of the current botnet measurement
botherders can modify the C&C server to discard the techniques work on specific botnet C&C
responses of the users and commands. In addition, communication protocols and structure. Consequently,
botherders can set the user mode of the bots to be as botnets change their C&C communication
invisible: the status messages of the bots are then not architecture, these methods are rendered ineffective.
visible for all members of a channel. In such According to our comparison, the most recent botnet
situations, the size of these botnets is hard to estimate measurement technique, based on infiltration as well as
[2]. In Passive technique there are several additional DNS based botnet measurement approach, can measure
issues that complicate the task of counting botnet real- world botnets regardless of its protocol and
members. Temporary bot migration and bot cloning are structure with a very low false positive rate. Hence,
major contributors to this effect. In several occasions, developing active techniques based on infiltration and
botmasters temporarily migrate their bots from one redirection for botnet measurement may be the most
botnet to another. In cloning, botmasters command bots promising approach to combat botnet.
to create copies of themselves and join a new channel
on the same server, or to connect to a different server REFERENCES
altogether. Generally, there are two types of cloning:
[1] M. A. R. J. Z. Fabian and M.A. Terzis, "A multifaceted
clone flooding, in which bots create a large number of approach to understanding the botnet phenomenon," in
instances to overwhelm a target IRC server; and; Proceedings of the 2006 ACM SIGCOMM Internet
normal cloning events in which botmasters command Measurement Conference (IMC), 2006.
their bots to create a new IRC connection and join [2] J. Zhuge, T. Holz, X. Han, J. Guo, and W. Zou, Characterizing
the IRC- based botnet phenomenon: University of
another channel on the same server or on a different Mannheim/Institute of Computer Science and Technology
server. Beijing, China, 2007.
[3] M. A. R. J. Z. Fabian and M.A. Terzis, "My botnet is bigger [10] H. Jun, L. Zhitang, Y. Dezhong, and Y. Junfeng,
than yours (maybe, better than yours): why size estimates "Measuring Botnet Size by Using URL and Collaborative
remain challenging," in Proc. of the 1st Workshop on Hot MailServers," in Networking and Services, 2009. ICNS '09.
Topics in Understanding Botnets (HotBots 2007), 2007. Fifth International Conference on, 2009, pp.161-164.
[4] T. Holz, M. Steiner, F. Dahl, E. Biersack, and F. Freiling, [11] A. Krishnamurthy, B. Plattner, and R. Weaver, "A Probabilistic
"Measurements and mitigation of peer-to-peer-based botnets: a Population Study of the Conficker-C Botnet," in Passive and
case study on storm worm," in Proceedings of the 1st Usenix Active Measurement. vol. 6032: Springer Berlin Heidelberg,
Workshop on Large-Scale Exploits and Emergent Threats, 2009, pp. 181-190.
2008, pp. 1-9. [12] M. Feily, A. Shahrestani, and S. Ramadass, "A Survey of
[5] I. Castle and E. Buckley, "The Automatic Discovery, Botnet and Botnet Detection," in Emerging Security
Identification and Measurement of Botnets," in Emerging Information, Systems and Technologies, 2009. SECURWARE
Security Information, Systems and Technologies, 2008. '09. Third International Conference on, 2009, pp. 268-273.
SECURWARE '08. Second International Conference on, 2008, [13] Z. Li, J. Hu, Z. B. Hu, B. Wang, L. Tang, and X. Yi,
pp. 127-132. "Measuring the botnet using the second character of bots,"
[6] E. Van Ruitenbeek and W. H. Sanders, "Modeling Peer-to-Peer Journal of Networks, vol. 5, pp. 98-105, 2010.
Botnets," in Quantitative Evaluation of Systems, 2008. QEST [14] R. Weaver, "A probabilistic population study of the Conficker-C
'08. Fifth International Conference on, 2008, pp. 307-316. botnet," in Passive and Active Measurement, 2010, pp. 181-
[7] W. Binbin, L. Zhitang, T. Hao, and M. Jie, "Measuring Peer-to- 190.
Peer Botnets Using Control Flow Stability," in Availability, [15] D. Plohmann, E. Gerhards, and F. Leder, Botnets: Detection,
Reliability and Security, 2009. ARES '09. International Measurement, Disinfection & Defence, 2011.
Conference on, Wuhan, China, 2009, pp. 663-669. [16] H. Xin, M. Knysz, and K. G. Shin, "Measurement and analysis
[8] W. Binbin, L. Zhitang, T. Hao, H. Zhengbing, and H. Jun, of global IP-usage patterns of fast-flux botnets," in 2011
"Actively Measuring Bots in Peer-to-Peer Networks," in Proceedings IEEE INFOCOM, 2011 Proceedings IEEE, 2011,
Networks Security, Wireless Communications and Trusted pp. 2633-2641.
Computing, 2009. NSWCTC '09. International Conference on, [17] L. Shangdong, G. Jian, Y. Wang, and A. Jakalan, "A Survey of
2009, pp. 603-607. Botnet Size Measurement," in Networking and Distributed
[9] B. B. H. Kang, E. Chan-Tin, C. P. Lee, J. Tyra, H. J. Kang, C. Computing (ICNDC), 2011 Second International Conference
Nunnery, Z. Wadler, G. Sinclair, N. Hopper, and D. Dagon, on, 2011, pp. 36-40.
"Towards complete node enumeration in a peer-to-peer botnet,"
in Proceedings of the 4th International Symposium on
Information, Computer, and Communications Security, 2009,
pp. 23-34.
Abstract—This paper studies discovery of resources distributed database. The problem is complicated by the
on virtual organization using flooding algorithm. The fact that some resource information (e.g., CPU load or
main problem arises when we have to find the resource in available storage) changes dynamically. Resource
grid in a lesser time. Here match making rules are taken discovery techniques maintain their source attribute and
to identify each resource type and distance travelled in
hops. If resource matches in lesser time then it is
status information in a distributed database and differ in
immediately allocated to the user. We make use of router the way they update, organize, or maintain the
information which is updated as the number increase4 or distributed database.
decreases; this increases the efficiency of memory
utilization on the router.
Keywords: Resource Discovery, Virtual organization,
Reliability value, Flooding algorithm.
I. INTRODUCTION
Grid computing provides an effective infrastructure
for massive computation among flexible and dynamic
collection of divided system for resource discovery.
Grid system is a large scale of distributed environment
which provides a high number of powerful resources to
its users. It provides many types of resources such as
computing resources (CPU cycles, storage, network
bandwidth and memory etc) and services (access to
specific data, shared software etc) [1].
Resource Discovery is systematic process of
Fig. 1: Grid Resource Discovery [12]
determining which grid resource is the best candidate to
complete a job with trade-offs like The challenge is to devise highly distributed
1. In shortest amount of time discovery techniques that are fault tolerant and highly
2. With most efficient use of resources scalable. Matching the needs of an application with
3. At minimum cost. available resources is one of the basic and key aspects
Resource Discovery schema maintains and queries of a Grid system [4].
a resource database known as “status” database. This
database is maintained network wide to fulfill the client II. RELATED WORK
request information service. Discovery is initiated by a
network application to find suitable resources within the In this section, research work concerning the grid
Grid. For this it has to interact with the resources resource discovery problem in general, as well as
individually through agents or it maintains the implementation of trust mechanisms is presented.
information about the resources in databases. Though significant work has been done towards the
The grid resource discovery problem [3] can be direction of trusted grid computing [5] focusing to the
defined as the problem of matching a query for security implications that could arise in such systems,
resources, described in terms of required characteristics, resource discovery in grid system is limited. In this
to a set of resources that meet the expressed section we presented related work regarding resource
requirements. The problem is complicated by the fact discovery implementations.
that some resource information (e.g., CPU load or The amount of communication required by the
available storage) changes dynamically. Resource algorithm is measured by:
discovery techniques maintain their source attribute and a) The pointer communication complexity defined
status information in a distributed database and differ in as the number of pointers exchanged during
the way they update, organize, or maintain the the course of the algorithm.
Abstract—The efficient node-energy utilization is one Sivalingam K., 2001, Manjeshwar A. and. Agrawal
of important factors in wireless sensor network, research D.,2001). For these sensor network applications, most
on wireless sensor network has often assumed research has discussed problems by the deployment of
homogenous node which are equipped with the same large number of low-cost homogeneous devices.
amount of energy and as a result, they cannot take
advantage of heterogeneity. This leads to the research in
However, it is often feasible to consider the deployment
heterogeneous network in which fraction of nodes are of heterogeneous devices with different capabilities
powerful (with more energy), more suitable to become a Advances in wireless sensor network technologies have
cluster head. Intuitively, powerful nodes have to become made it possible to develop low-cost, low-powered tiny
cluster head more often than the normal nodes. In this sensor nodes.
paper we proposed a cluster based routing algorithm to
extend the lifetime of the networks and to maintain the
balanced energy consumption of nodes, we analyzed a
heterogeneous network with three types of nodes having
different initial energy levels. The performance of the
proposed protocol has been examined and evaluated in
MATLAB simulator. As a result of simulation, we have
confirmed that our proposed algorithm shows better
performance in terms of lifetime than LEACH, without
other critical overheads and performance degradation.
Keywords: Wireless Sensor Network (WSN), Routing
Protocols, Heterogeneous, Lifetime, Clustering, Fig. 1: Architecture of Sensor Network
Homogeneous.
The major issues stemming from these studies are
I. INTRODUCTION protocol design in regards to battery energy efficiency,
localization scheme, synchronization, data aggregation
A wireless network can be defined as a network and security technologies for wireless sensor networks.
consists of low size and low complex devices are called In particular, researchers have shown great interest in
sensor nodes which can sense surrounding things and the routing protocols in the network layer, which
after gathering information from that monitored considers self-organization capabilities, limited battery
surrounding fields and communicates through wireless power, and data aggregation schemes (K. Akkaya and
links. M. Younis, 2005, Q. Jiang and D. Manivannan,
Wireless communication technologies continue to 2004).The concept of cluster based routing is also
grow in diverse areas to provide new and better utilized to perform energy efficient routing in WSNs.
opportunities for general business systems; the Author(C. Siva Ram Murthy and B.S Manoj,2004)
advances in micro-electro mechanical systems have shows the most important features of cluster-based
made building such kind of sensors a possibility. The architectures over ad hoc and sensor networks.
Wireless Sensor Networks (WSNs) consist of large Researchers generally assume that the nodes in wireless
number of low cost devices to gather information from sensor networks are homogeneous. But in reality,
the diverse kinds of physical phenomenon. The sensors homogeneous sensor networks hardly exist. In
can monitor various entities such as: temperature, heterogeneous sensor networks, few nodes are having
pressure, humidity, salinity, metallic objects, and comparatively more energy to perform data filtering,
mobility; this monitoring capability can be effectively fusion and transport. This leads to the research on
used in commercial, military, and environmental heterogeneous networks to prolong the lifetime and
applications (Heinzelman W., Chandrakasan A., and reliability of the network.
Balakrishnan H.2000, Lindsay S., Raghavendra C., and
For heterogeneous WSNs, a very critical task for II. RELATED STUDY
clustering protocols is to select the cluster head so that
In a cluster based routing protocol, sensor nodes
least energy is consumed, and hence prolong the
are partitioned into many groups (clusters), which
lifetime. Clustering algorithms can be classified based
integrates information data collected from sensor nodes
on two main criterions: according to the energy
and transmits them to the sink node of the network. In
efficiency and stability. Selection of cluster head in
each of the many clusters in this network resides a
energy efficient techniques generally depends on the
cluster head which collects data from sensor nodes
initial energy, residual energy, average energy of the
within its group, completes data aggregation, and sends
network, or energy consumption rate or combination of
them to the sink node of the network. Such data
these. Some algorithm for clustering: EEHC (Dilip
aggregation can reduce the consumption of node energy
Kumar a,*, Trilok C. Aseri b, 1, R.B. Patel, 2010),
and the transmission delay as compared to multi-hop
DEEC (Vivek Katiyar, Narottam Chand, Surender Soni,
routing protocols (J. Ibriq and I. mahgoub, 2004).
2010), DBEC (Changmin Duan; Hong Fan, 2007) and
In recent years, there has been a growing interest in
CBSD (R.S. Marin-Perianu, J. Scholten, P.J.M.
WSNs. One of the major issues in wireless sensor
Havinga And P.H. Hartel, 2010).All the recent research
network is developing an energy-efficient routing
work focuses on sensor networks that consist of
protocol. Since the sensor nodes have limited available
identical sensors with equal capacity in terms of
power, energy conservation is a critical issue in wireless
sensing, computation, communication, and power. The
sensor network for nodes and network life. The issue of
possibility of working with more than one type of
heterogeneity (in terms of energy) of nodes is addressed
sensors within a same network is mentioned in
in (Ye M., Li C., Chen G., and Wu J., 2005). In
(Pottie G. and Kaiser W., 2000). We have observed that
(Smaragdakis G., Matta I., and Bestavros, 2004), the
the manufacturing of a sensor is generally application-
proposed protocol is based on random selection of
specific. Different special purpose sensors can be used
cluster-heads weighted according to the remaining node
to form a single sensor network to perform more
energy. This approach addresses the problem of varying
comprehensive tasks, e.g., some sensors collect image
energy levels and consumption rates but still assumes
data, some sensors collect audio signal, some sensors
that the sink can be reached directly by all the nodes.
have more processing capability, some sensors have
There have been many studies presented for
more power, and so on. This results in a heterogeneous
configuration and operation algorithms of a cluster
sensor network that can have a variety of compositions
based network topology in ad hoc networks.
of sensors. Many organizational and communication
Advantages to a clustering network are the reduction of
issues arise with such a structure.
an overhead of routing establishment, minimization of
In this paper, we have examined one of the
the size of the routing table, and stabilization of the
simplest heterogeneous scenarios in which sensors are
network topology. The clustering network can make
equipped with different battery power in an event-
driven sensor network. In particular, we have resource management and bandwidth allocation more
considered a field randomly deployed with sensors that efficient and make node positioning management and
gather data and transmit it back to a base station. Such a transmitting power management possible.
scenario is motivated by applications in which data is In addition, the clustering scheme in a wireless
desired from a hostile environment, such as a volcano sensor network enables an aggregate data of cluster
or a swamp, where sensors are likely to be deployed in member nodes at the cluster head and can easily
an unmanned manner. Under such situations sensing provide the network scalability due to node increase.
data will be collected, aggregated, analyzed and Consequently, in a huge sensor network that is several
transmitted to a more accessible location. The main hundred ~ hundred thousand times bigger than ad hoc
issue of our interest is to maximize the lifetime of a networks; the clustering based routing algorithm is a
sensor network for a given amount of energy, or possible approach to maintain network configuration
equivalently, and to retrieve the same data using the management and to make data aggregation. In the
least amount of energy also the proposed algorithm clustering algorithm, all nodes in a sensor network can
extends the survival time of sensor networks by become a cluster head but must belong to only one
properly choosing cluster heads with consideration to cluster. The algorithm should minimize the overhead of
node residual energy. clustering setup messages and establishing times.
The rest of the paper is organized as follows. Next, Additionally, the algorithm must maintain a stable
the related work is discussed. Then we show a paradigm network configuration, routing, network efficiency,
of heterogeneous wireless sensor networks and also with a minimization of energy Consumption
describe the network model. Then we describe the basic (J. Ibriq and I. mahgoub, 2004).
system model and proposed work. Lastly we present
simulation results and conclude the paper.
• In the set-up phase keeping the base station position at the center denoted
1. Each node generates a random probability by ‘x’.We denote the normal node with ‘o’, an
(p) at the beginning of a new round and advanced node with ‘+’ and a super node with
computes the threshold value (T (n)) with ‘^’deployment of the sensor node is shown in figure
the use of equation (1). 4.Our simulation model uses the parameters as shown
If r = 1 (i.e. the first round), let Emax of all in table 1
nodes be 1. TABLE 1: SIMULATION PARAMETERS
2. In case of p < Pt, the node is selected as a
cluster head. Simulation Parameter Value
Network area 200m *200m
3. A selected cluster head broadcasts an No of nodes 100
advertised message over neighbor nodes. Initial energy 0.5J
4. The neighbor nodes collect advertised BS position 100m * 100m
messages during a given time interval and Eelec 50nJ/bit
then send a “join REQ” message to the Etx=ERX 50nJ/bit
εfs 10pJ/bits/m2
nearest cluster head. εmp 0.0013pJ/bit/m4
5. The cluster head receives the “join-REQ” do=sqrt(εfs / εmp)
messages and builds a cluster member list EDA 5nJ/bit
schedule. Then broadcast them over Packet size 4000bits
neighbor nodes. β 2
α 1
6. The member node receives and save the
p 0.5
message for data transfer.
• In the steady state phase,
1. After the cluster selection process
completes, each member sends data and
its residual energy information to them
cluster head
2. The cluster head maintains residual energy
information of member nodes.
• In the Pre-setup phase,
1. Before the last frame of a round
completes, the cluster head sends BS the
maximum residual energy value of nodes
belong to its own cluster.
2. BS collects all maximum residual energy
values from cluster heads, finds the Fig. 4: Simulation Result of Sensor Network
maximum residual energy value (Emax) B. Simulation Results
of the network, and sends Emax back to
cluster heads. 1) Network lifetime when first node dead
3. The cluster head broadcasts Emax over In the random selection method cluster head
cluster nodes. selection in each round is done on the basis of 1/p, not
4. Each node save the value of Emax for the on the basis of residual energy. In this we observe the
next computation of T (n) and the current result of the first node dead, because dead nodes are the
round is terminated reason to affect the lifetime of the network or drag
towards early dead of all node,the base station position
VI. PERFORMANCE EVALUATION placed at the (100,200) with 4000 packet size.
The performance analysis of above routing Simulation is taken for 2000 rounds.
protocol is evaluated with the MATLAB (www. TABLE 2: NETWORK LIFETIME (FIRST NODE DEAD)
Mathworks.com). Then the protocol is compared to the
Simulation Rounds LEACH Proposed Protocol
heterogeneous LEACH algorithm (in which cluster 1 438 635
selection is done by random selection) in terms of the 2 501 661
network lifetime. 3 462 640
4 560 723
A. Simulation Parameter 5 478 678
6 456 639
In this simulation, our experiment model performed 7 467 712
on 100 nodes which were randomly deployed and 8 532 698
distributed in a 100×100 square meter area. Initially 9 521 694
10 448 742
3000 1200
2500
1000
No of Rounds
2000
800
No of Rounds
1500 4000
LEACH
1000 600
Proposed 8000
500 400
12000
0 200
1 2 3 4 5 6 7 8 9 10
Simulation Rounds 0
Fig. 5: Network Lifetime (First Node Dead) 1 Leach 2 3 Proposed
2) Network lifetime when last node dead Fig. 7: Network Lifetime (Different Packet Size)
Similarly when all nodes are dead in the network, 4) Residual energy of the network
the lifespan of network is over, less the round number We simulated the proposed method to get the value
implies lesser lifetime of network. Figure 6 shows the of residual energy left for the network at differ rounds
result, Simulation is taken for 2000 rounds. and we compared with the LEACH protocol and the
TABLE 3: NETWORK LIFETIME (ALL NODE DEAD) result are shown in figure 8,and the result shows that
Simulation Rounds LEACH Proposed
our proposed protocol have more residual energy for the
1 1813 2314 network. Keeping the base station position at (100,200)
2 2145 2456 with packet size 4000.
3 1987 2221
TABLE 5: RESIDUAL ENERGY IN JOULES FOR DIFFERENT ROUNDS
4 1869 2532
5 2145 2654 Round Number LEACH Proposed
6 1854 2433 500 22.595 34.567
7 1865 2563 1000 14.097 25.121
8 1923 2321 1500 5.2684 16.476
9 2103 2422 2000 2.1902 9.4093
10 1821 2466 2500 1.5828 4.8712
3000 0.5632 3.0321
3000
3500 0.0666 0.9765
2500 4000 0 0.5521
4500 0 0
No of Rounds
2000
1500 40
LEACH 35
Residule energy
1000 30
Proposed 25
500 20
15
0 10 LEACH
5
1 2 3 4 5 6 7 8 9 10 0 Proposed
Simulation Rounds
500
1000
1500
2000
2500
3000
3500
4000
4500
residual energy during cluster head selection processing [8] Vivek Katiyar, Narottam Chand, Surender Soni,” Clustering
improves the network performance.By considering the Algorithms for Heterogeneous Wireless Sensor Network: A
Survey “, International Journal Of Applied Engineering
base station position, packet size, our protocol clearly Research, Dindigul Volume 1, No 2, 2010.
makes the network lifetime much longer without other [9] Changmin Duan; Hong Fan, “A Distributed Energy Balance
critical overhead and performance degradation. Clustering Protocol for Heterogeneous Wireless Sensor
Networks”, Wireless Communications, Networking and Mobile
Computing, 2007.
REFERENCES [10] R.S. Marin-Perianu, J. Scholten, P.J.M. Havinga And P.H.
[1] Heinzelman W., Chandrakasan A., and Balakrishnan H., Hartel Cluster -based service discovery for heterogeneous
“Application-Specific Protocol Architectures for Wireless wireless sensor networks International Journal of Parallel,
Networks,” PhD Thesis, Massachusetts Institute of Technology, Emergent and Distributed Systems, Vol., No., 1–35.
2000. [11] Pottie G. and Kaiser W., “Wireless Integrated Network
[2] Lindsay S., Raghavendra C., and Sivalingam K., “Data Sensors,” Communications of the ACM, vol. 43, no. 5, pp. 51-
Gathering in Sensor Networks Using the Energy Delay Metric,” 58, 2000.
in Proceedings of the 15thInternational Parallel and Distributed [12] J. Ibriq and I. mahgoub, "Cluster-Based Routing in Wireless
Processing Symposium, 2001. Sensor Networks: Issues and Challenges", Proceedings of the
[3] Manjeshwar A. and. Agrawal D., “TEEN: A Routing Protocol 2004 Symposium on Performance Evaluation of Computer
for Enhanced Efficiency in Wireless Sensor Networks,” in Telecommunication Systems, pp.759-766, Jul. 2004.
Proceedings of 1st International Workshop on Parallel and [13] Ye M., Li C., Chen G., and Wu J., “EECS: An Energy Efficient
Distributed Computing Issues in Wireless Networks and Mobile Clustering Scheme in Wireless Sensor Networks,” in
Computing, San Francisco, 2001. Proceedings of 24th IEEE International Performance,
[4] C. Siva Ram Murthy and B.S Manoj, Ad Hoc Wireless Computing, and Communications Conference (IPCCC’2005),
Networks Architectures and Protocols. Prentice Hall pp. 535-540, 2005.
Communication Engineering and Emerging Technologies [14] Smaragdakis G., Matta I., and Bestavros A., “SEP: A Stable
Series, 2004. Election Protocol for Clustered Heterogeneous Wireless Sensor
[5] K. Akkaya and M. Younis, "A Survey of RoutingProtocols in Networks,” in Proceedings of 2nd International Workshop on
Wireless Sensor Networks, " in the Elsevier Ad Hoc Network Sensor and Actor Network Protocols and Applications (SANPA’
Journal, Vol 3/3, pp. 325-349, 2005. 2004), 2004
[6] Q. Jiang and D. Manivannan, “Routing protocols for sensor [15] www. Mathworks.com
networks,” Proceedings of CCNC 2004, pp.93-98, Jan. 2004.
[7] Dilip Kumar a,*, Trilok C. Aseri b, 1, R.B. Patel “EEHC:
Energy efficient heterogeneous clustered scheme for wireless
sensor networks in “Computer Communications” 32 (2009)
662–667.
Abstract—Next generation networks will certainly 3. Share the cost of managing the networks (as
face requesting access from different parts of the IBM’s Autonomic Computing)[1].
network.The heterogeneity of communication and
4. Support mobile, personalized Services
application softwares changing situations in the
environment, from the users, the operators, the business 5. Improve efficiencies through context
requirements as well as the technologies. Users will be awareness, etc.
more and more mobile, protocols, etc. will increase and 6. End-user is becoming a resource-powerful
render the network more complex to manage. Autonomic engine.
networking aims to design networks that are capable to 7. Contributes its resources to the network
self-manage, while optimizing their configurations and 8. The autonomic, powerful end-user is
interactions to the changing needs of the users and the
environment.The present paper deals with overview of
becoming the network.
autonomic communication and its role in digital global 9. The network will be made up of zillions of
world. autonomic nodes.
actions may lead the system/subsystem to alter its own limitation providing the system with the ability to
state and/or influence changes in other elements of the represent and reason about, and to compare and
environment. The information pervaded environment as evaluate, possible situations that do not exist, or exist
a whole becomes more resilient as changes are made to but are not known, either because they are future
reduce or help to eliminate the impact of failing possibilities, or because they are remote or hypothetical
components. possibilities or because they might have occurred in the
past [5 ]. Finally, the reflective process enables the
D. Self-optimizing cognitive elements to monitor and control their own
This feature refers to the ability of the information- progress to adapt their future behavior in response to
pervaded environment to efficiently maximize resource previous performance. Statistics, or executions history,
allocation and utilization to meet end-users needs with are an important parameter in the decision-making
a minimal human intervention. In the near term, self- process.The architecture aims to reproduce the
optimization primarily addresses the complexity of behavior of the modeled system (i.e. human nervous
managing system performance. system) at different level of time (short term, long term
reactions) to exhibit a robust behavior to errors and
E. Self-protecting unknown/unexpected events.
The goal of self-protecting environments is to
provide the right information to the right users at the
right time through actions that grant access based on
the users role and predefined privileges. A self
protecting information-pervaded environment can
detect hostile or intrusive behavior as it occurs and take
autonomous actions to make itself less vulnerable to
unauthorized access and use, viruses, denial-of-service
attacks and general failures.
F. Self- and Context-aware
These features refer to perception and cognitive
reaction to an event or more generically to a condition,
relevant to the same intelligent node or, respectively to
the environment. Context-awareness is a foundation for
the rest of the operational features: self-configuring, Fig. 2: DARPA Cognitive Architecture
self-healing, self-optimizing, and self-protecting.
VIII. INFORMED DECISION IN WIRELESS ECOSYSTEM
VII. COGNITIVE RADIO IN AUTONOMIC DOMAIN
An extended model for autonomic computing and
Cognitive architectures are characterized by certain autonomic communications is proposed and depicted in
properties: the cognitive behavior should not be figure 3.
implemented partly but it should concern the complete
system implementation of various aspects of the
cognitive behavior as well as the complete system
(Holistic view of the cognitive system).The system
aims to learn and adapt its future reactions based on
statistics about the previous executions. DARPA
proposes an architecture for cognitive nodes that
introduces 3 processes related to cognitive behaviors:
reactive, deliberative and reflective (meta
management)reasoning process.
The “reactive” process provides the cognitive
network with capabilities that allows reacting
automatically to some events or perception of the Fig. 3: Model of Autonomic Computing and Autonomic
environment. However, that sense of “reactive” Communication
excludes any possibility to take into account future
The model introduces some new features. The first
possibilities, hypotheses about what might have been
is the functionality for sensing,assessing risks and
the case, or formulate hypotheses about what exists in
determining behavior[3].The second issue is the
some part of the system that is not currently being
computation of fitness function.
perceived. Deliberative process permits to leverage this
Abstract—In the software testing a real way of Data Generation (ATDG) Technique is a technique
measuring effective testing is that the system passes an used for automatic test data generation for SUT.
adequate suite of test cases, and then it must be correct or Figure 1 models a typical test data generator
dependable. But that’s impossible because the adequacy of system which consists of three parts: program analyzer,
test suites is provably undecidable. So we’ll have to settle
adequacy Design rules to highlight inadequacy of test
path selector and test data generator. The source code is
suites. Design rules do not guarantee good designs. Good run through a program analyzer, which produces the
design depends on talented, creative, disciplined designers necessary data used by the path selector and the test
and that can be the best test suite set to fulfill adequacy data generator. The selector inspects the program data
criteria. in order to find suitable paths. Suitable for instance
mean paths leading to high code coverage.
Keywords: adequacy criteria, automatic test case
generation, test data generation.
Abbreviations: GA- Genetic Algorithm, SUT- Software
under Testing
I. INTRODUCTION
If we talk about testing then the role of test case
designing is very important. A Test case is a set of
inputs, execution conditions, and a fail/pass criterion. A
requirement to be satisfied by one or more test cases. A
test suite is created by such test cases. Test obligation is
a partial test case specification, requiring some property
deemed important to thorough testing.
If a test suite fails to satisfy some criterion, the
obligation that has not been satisfied may provide some
useful information about improving the test suite. If a
test suite satisfies all the obligations by all the criteria,
we do not know definitively that it is an effective test
suite, but we have some evidence of its thoroughness.
II. TEST DATA GENERATION Fig. 1: Architecture of a Test Data Generator System
Software testing accounts for 50% of the total cost The paths are then given as argument to the test
of software development. This cost could be reduced if data generator which derives input values that exercise
the process of testing is automated. One way to do this the given paths. The generator may provide the selector
would be to generate input data to the program to be with feedback such as information concerning
tested-program-based test data generation. infeasible paths.
Through the years a number of different methods A control flow graph, or just flow graph when
for generating test data have been presented. In 1996 context is clear, is a graphical representation of a
Ferguson and Korel divided these methods in three program. There exist many different definitions on
classes: random, path-oriented, and goal-oriented test control flow graphs throughout the literature. The
data generation. This is the most appropriate definition used here has been inspired by Beizer as well
classification in terms of test data generation, although as Korel. Let’s consider a program to understand.
the problem of path selection is not considered int triType( int a, int b, int c)
separately. The selection of a path can largely affect the {
whole process of test data generation. 1 int type = PLAIN;
Test data is basically the input needed for 1 if(a<b)
executing the test code, which has been written to test 2 swap(a,b);
the ‘software under test’ (SUT) and Automated Test 3 if(a<c)
branch, path, or definition–use pair in the program chromosomes iteratively evolves to one in which
under test. The genetic algorithm conducts its search by candidates satisfy some termination criteria or, as is
constructing new test data from previously generated also in our case, fail to make any progress. Each
test data that are evaluated as good candidates. The iteration step is also called a generation.
algorithm evaluates the candidate test data, and hence Each generation may be viewed as a two stage
guides the direction of the search, using the program’s process [Whi94]. Beginning with the current
control dependence graph. population, selection is applied to create an intermediate
A GA operates on strings of digits called population and then crossover and mutation are applied
chromosomes, each digit that makes up the to create the next population. The most common
chromosome is called a gene, and a collection of such selection scheme is the roulette-wheel selection in
chromosomes makes up a population. Each which each chromosome is allocated a wheel slot of
chromosome has a fitness value associated with it, and size in proportion to its fitness. By repeatedly spinning
this fitness value determines the probability of survival the wheel, individual chromosomes are chosen using
of an individual chromosome in the next generation. “stochastic sampling with replacement” to construct the
After the next generation is created a percentage of the intermediate population. Other selection schemes also
chromosomes are crossed and a small percentage of the exist [Gol89], for instance the elitist scheme in which
genes are mutated. the fittest chromosomes survive from one population to
A basic algorithm for a GA is as follows: the other.
1. Generate a random population of Selection Recombination
chromosomes. (Duplication) (Crossover)
2. Calculate the fitness of each member of the
population. String 1 String 1 Offspring-A (1 X 2)
3. While best chromosome (highest fitness) is String 2 String 2 Offspring-B (1 X 2)
below some threshold or the number of String 3 String 2 Offspring-A (2 X 4)
generations is below some limit. String 4 String 4 Offspring-B (2 X 4)
• Generate next population (reproduction)
allowing greater chance of more fit
population members to survive.
• Crossover a percentage of population
members to create two new members.
• Mutate a percentage of genes in the
population.
Current Next
• Recalculate fitness of the population. Generation t
Intermediate
Generation t + 1
Generation t
4. Finish.
Fig. 5: One Generation is Broken Down into a Selection Phase and
While the algorithm begins with a random Recombination Phase
population, it is the iterative process of reproduction,
crossover, and mutation that gives the process a This figure shows strings being assigned into
structure. A chromosome on its own may have a poor adjacent slots during selection. In fact, they can be
fit or low fitness, but it may contain some assigned slots randomly in order to shuffle the
characteristics which when combined with another intermediate population. Mutation (not shown) can be
chromosome (crossover) or adjusted slightly (mutation) applied after crossover.
will create an offspring with a much higher fitness.
A genetic algorithm execution begins with a
random initial population of candidate solutions {is} to
an objective function f(s). Each candidate solution is is
generally a vector of parameters to the function f(s) and
usually appears as an encoded binary string (or bit
string) called a chromosome. An encoded parameter is
referred to as a gene, where the parameter’s values are
the gene’s alleles. The number of parameters and the
size of the bit encoding of each parameter defines the
length of the chromosome. In this part a chromosome
represents an encoding of a test case.
After selection & crossover recombination is
After creating the initial population, each
applied to randomly paired strings with a probability.
chromosome is evaluated and assigned a fitness value.
Amongst the various crossover schemes. The one point,
Evaluation is based on a fitness function that is problem
two point and the uniform crossover schemes are few
dependent. From this initial selection, the population of
popular schemes [Gol89, Whi94]. In the one point case
a crossover point is identified in the chromosome bit “Testing never proves the absence of faults, it only
string at random and the portions of chromosomes shows their presence.”
following the crossover point. In the paired crossover Code Coverage is a way to try to cover more of the
chromosomes, are interchanged, while in the two point testing problem space so that we come closer to proving
case two crossover points are identified in the bit string the absence of faults, or at least the absence of a certain
and the middle portions are exchanged in the paired class of faults. The code coverage is just one weapon in
chromosomes. In the case of uniform crossover each bit the software engineer's testing arsenal.
in the string is exchanged according to a predefined The Statement Coverage is the most basic form of
crossover probability. code coverage. In software testing practice, testers are
In addition to crossover, mutation is used to often required to generate test cases to execute every
prevent permanent loss of any particular bit or allele. As statement in the program at least once. A test case is an
an operator, mutation application also introduces input on which the program under test is executed
genetic diversity. Mutation results in the flipping of bits during testing. A test set is a set of test cases for testing
in a chromosome according to a mutation probability a program. The requirement of executing all the
which is generally kept very low. statements in the program under test is an adequacy
The chromosome length, population size, and the criterion. A test set that satisfies this requirement is
various probability values in a GA application are considered to be adequate according to the statement
referred to as the GA parameters. Selection, crossover, coverage criterion. Sometimes the percentage of
mutation are also referred to as the GA operators. executed statements is calculated to indicate how
adequately the testing has been performed. The
IV. ADEQUECY CRITERIAN percentage of the statements exercised by testing is a
measurement of the adequacy.
Adequacy criterion is a predicate that is true
A statement is covered if it is executed. A
(satisfied) or false (not satisfied) of a 〈program, test statement does not necessarily correspond to a line of
suite〉 pair. It is the set of test obligations. A test suite code multiple statements on a single line.
satisfies an adequacy criterion if all the tests succeed Where there are sequences of statements without
(pass) every test obligation in the criterion is satisfied branches it is not necessary to count the execution of
by at least one of the test cases in the test suite. every statement, just one will suffice. But people often
Adequacy criteria create test cases to cover faults like the count of every line to be reported especially in
that could possibly occur in the software, introduce summary statistics.
mutations in the code. The statement coverage It can be quite difficult to achieve 100% statement
adequacy criterion is satisfied by test suite S for coverage. There may be sections of code designed to
program P if each executable statement in P is executed deal with error conditions, or rarely occurring events
by at least one test case in S, and the outcome of each such as a signal received during a certain section of
test execution was “pass”. Sometimes no test suite can code. There may also be code that should never be
satisfy a criterion for a given program. executed:
if (a==0) int a=10;
{ if (a<0)
c=b/a; print "This should never happen!";
throw new Logic Error (“a must be non zero Statement coverage, or something very similar, can
here!”) also be called statement execution, line, block, basic
} block or segment coverage.
Simply put, code coverage is a way of ensuring that The goal of Branch Coverage is to ensure that
your tests are actually testing your code. When you run whenever a program can jump, it jumps to all possible
your tests you are presumably checking that you are destinations. Similarly the branch coverage criterion
getting the expected results. Code coverage will tell you requires that all control transfers in the program under
how much of your code you exercised by running the test are exercised during testing. The percentage of the
test. There are a number of criteria that can be used to control transfers executed during testing is a
determine how well your tests exercise your code. measurement of test adequacy. The simplest example is
a complete if statement:
V. COVERAGE MATRICS if (x)
A number of different metrics are used determine print "it will be execute";
how well exercised the code is ? The most simple is else
statement coverage, which simply tells you whether you print "it will not be execute";
exercised the statements in your code. When working Full coverage is achieved here only if x is true on
with code coverage, we should keep in mind that one occasion and false on another.
Achieving full branch coverage will protect against means that the second term will not be evaluated if the
errors in which some requirements are not met in a value of the first term has already determined the value
certain branch. For example: of the whole expression. For example, when using the ||
if (x!=0) operator the second term is never evaluated if the first
{ evaluates to true. This means that for full condition
print "x is nonzero"; coverage there are only three combinations of values to
z=y/x; cover instead of four.
} Condition coverage gets complicated and difficult
else to achieve, as the expression gets complicated. For this
{ reason there are a number of different ways of reporting
print "x is zero"; condition coverage which try to ensure that the most
} important combinations are covered without worrying
This code will fail if x is false. In such a simple about less important combinations. Expressions which
example statement coverage is as powerful, but branch are not part of a branching construct should also be
coverage should also allow for the case where the else covered:
part is missing, and in languages which support the z = x || y;
construct, switch statements should be catered for: Condition coverage is also known as expression,
if (x!=0) condition-decision and multiple decision coverage.
{ Mutation Adequacy is often aimed at detecting
print "x is nonzero"; faults in software. A way to measure how well this
z=y/x; objective has been achieved is to plant some artificial
} faults into the program and check if they are detected by
100% branch coverage implies 100% statement the test. A program with a planted fault is called a
coverage. Branch coverage is also called decision, arc mutant of the original program. If a mutant and the
or all edges coverage. original program produce different outputs on at least
There are classes of errors which branch coverage one test case, the fault is detected. In this case, we say
cannot detect, such as: that the mutant is dead or killed by the test set.
if (x) Otherwise, the mutant is still alive. The percentage of
{ dead mutants compared to the mutants that are not
print "it will be execute"; equivalent to the original program is an adequacy
} measurement, called the mutation score or mutation
if(y) adequacy [Budd et al. 1978; DeMillo et al. 1978;
{ Hamlet 1977].
print "it will also be execute"; From Good enough and Gerhart’s [1975, 1977]
} point of view, a software test adequacy criterion is a
100% branch coverage can be achieved by setting predicate that defines “what properties of a program
(x, y) to (1, 1) and then to (0, 0). But if we have (0, 1) must be exercised to constitute a ‘thorough’ test, i.e.,
then things go bang. The purpose of path coverage is to one whose successful execution implies no errors in a
ensure that all paths through the program are taken. tested program.” To guarantee the correctness of
adequately tested programs, they proposed reliability
Development efficiency = size of software/ Time and validity requirements of test criteria. Reliability
required requires that a test criterion always produce consistent
Time = Person days, Person Hours, Size= loc test results; that is, if the program tested successfully on
one test set that satisfies the criterion, then the program
Old generation languages may reduce productivity
also tested successfully on all test sets that satisfies the
as there are less chance of reusability. So for all above
criterion. Validity requires that the test always produce
mentioned adequacy criteria, objective of software’s a meaningful result; that is, for every error in a
has better chance of code coverage & will have better program, there exists a test set that satisfies the criterion
development efficiency due to reusability of code. and is capable of revealing the error. But it was soon
When a boolean expression is evaluated it can be recognized that there is no computable criterion that
useful to ensure that all the terms in the expression are satisfies the two requirements, and hence they are not
exercised. For example: practically applicable [Howden 1976]. Moreover, these
a if x || y; two requirements are not independent since a criterion
To achieve full condition coverage, this expression is either reliable or valid for any given software
should be evaluated with x and y set to each of the four [Weyuker and Ostrand 1980]. Since then, the focus of
combinations of values they can take. research seems to have shifted from seeking
In many software programming languages, most theoretically ideal criteria to the search for practically
boolean operators are short-circuiting operators. This applicable approximations.
Adequacy criteria provide a way to define a notion [3] [Duran and Natfos, 1984] Duran, J. and Natfos, S. (1984). An
evaluation of random testing. IEEE Trans-actions on Software
of “thoroughness” in a test suite.But they don’t offer Engineering, pages 438{444.
guarantees; more like design rules to highlight [4] [Gallagher and Narasimhan, 1997] Gallagher, M. J. and
inadequacy Narasimhan, V. L. (1997). Adtest: A test data generation suite
for ada software systems. IEEE TSE, 23(8):473{484.
[5] [Horgan et al., 1994] Horgan, J., London, S., and Lyu, M.
VI. CONCLUSION (1994). Achieving software quality with testing coverage
Adequacy criteria provide a way to define a notion measures. IEEE Computer, 27(9):60{69.
[6] [Korel, 1996] Korel, B. (1996). Automated test data generation
of “thoroughness” in a test suite. But they don’t offer for programs with procedures. In Proceedings of the 1996
guarantees; more like design rules to highlight International Symposium on Software Testing and Analysis,
inadequacy. pages 209{215. ACM Press.
The test criteria are a central problem of testing. [7] [Michael et al., 1997] Michael, C., McGraw, G., Schatz, M., and
Walton, C. (1997). Genetic algorithms for test data generation.
Numerous adequacy criteria have been proposed, In Proceedings of the Twelfth IEEE International Automated
analyzed and compared. A lot of research has been done Software Engineering Conference (ASE 97).
on the issue of evaluating and comparing criteria. The [8] [Ruchika Malhotra and Mohit Garg] Journal of Information
tendency is toward systematic approaches in testing Processing Systems, Vol.7, No.2, June 2011. An Adequacy
Based Test Data Generation. Technique Using Genetic
using adequacy criteria. Algorithms.
[9] [R. Mall. ]Fundamentals of Software Engineering. Prentice Hall
REFERENCES of India, 3rd edition. 2009
[10] [Elfriede Dustin] Effective Software Testing. Addison Wesley,
[1] [Chang et al., 1996] Chang, K., Cross, J., Carlisle, W., and Liao, 2002, ISBN 0-20179-429-2
S. (1996). A performance evaluation of heuristics-based test [11] [M. Tikir and J. Hollingsworth] “Efficient instrumentation for
case generation methods for software branch coverage. code coverage testing”, Int’l. Symp. on Software Testing and
International Journal of Software Engineering and Knowledge Analysis, Rome, Italy, 2002.
Engineering, 6(4):585{608. [12] [Offutt J., Liu S., Abdurazik A., and Ammann P] “Generating
[2] [Chilenski and Miller, 1994] Chilenski, J. and Miller, S. (1994). test data from state-based specifications”, Software Testing,
Applicability of modified condition/decision coverage to Verification and Reliability, Vol. 13, pp 25-53, 2003.
software testing. Software Engineering Journal, pages 193{200.
I. INTRODUCTION
Duobinary modulation techniques are known to
compress the optical spectrum, thereby facilitating the
tighter packing of channels into the EDFA gain
window. It has also been reported that the 2- level
variant of duobinary signaling [2, 3] almost eliminates
the impact of SBS since the optical carrier component is
suppressed [4]. Four-Wave-Mixing (FWM) is another
non linear effect that can limit the performance of
WDM systems [5, 6, 7, 8]. When high power optical
signal is launched into a fiber linearity of optical
response is lost. Four wave mixing is due to changes in Fig. 1: Duobinary Modulation Format Simulation Setup for 60 km
the refractive index with optical power called optical At output port of MUX a optical meter is used to
Kerr effect. In four wave mixing (FWM) effect, two co- measure output optical power from MUX and a optical
propagating wave produce two new optical sideband spectrum analyzer to view spectrum of signal which
wave at different frequencies. When new frequencies comes out from MUX. From output port of WDM
fall in the transmission window of original frequency it MUX signal travels to optical fiber of length 60km,
causes severe cross talk between channels propagating reference wavelength 1550nm, 3ps/nm/km dispersion
through an optical fiber. Degradation becomes very and core effective area that can be varied from 2.5 to
100 µm. Again at end of fiber another optical meter is
severe for large number of WDM channels with small
used to measure output optical power from fiber output
spacing. Optical duobinary modulation has attracted and a optical spectrum analyzer to view spectrum of
much attention as a transmission technique that can signal. At fiber end a WDM DEMUX is used having
mitigate fiber chromatic-dispersion effects in high same channel spacing as of MUX, the DEMUX has 100
capacity optical transmission system [9-11]. dB depth 15 GHz bandwidth and 2nd orber Bessel filter.
binary
-35
10
BER
-30
10
-25
10
-20
10
0 20 40 60 80 100
2
CORE EFFECTIVE AREA (µm )
(a)
REFRENCES
[1] P.L. Li, D.X. Huang, X.L. Zhang, J. Chen, and L.R. Huang,”
Theoretical Analysis of Tunable Wavelength Conversion Based
on FWM in a Semiconductor Fiber Ring Laser,” IEEE Journal
of quantum electronics, Vol. 41, No. 3, pp. 581-588, March
2005.
[2] T. Franck, P.B. Hansen, T.N. Nielsen, L. Eskildsen. Duobinary
Transmitter with Low Intersymbol Interference. IEEE Photon.
Technol. Lett., Vol. 10, No. 4, April 1998.
[3] T. Ono, Y. Yano, K. Fukuchi, T. Ito, H. Yamazaki, M.
Yamaguchi, K. Emura, “Characteristics of Optical Duobinary
Signals in Terabit/s Capacity, High-Spectral Efficiency WDM
Systems”, J. Lightwave Tech, Vol. 16, No. 5, May 1998.
[4] T. Franck, T.N. Nielsen, A. Stentz,.Experimental Verification of
SBS Suppression by Duobinary Modulation., ECOC 97,
September 1997.
[5] K. Inoue,. Influence of fiber FWM in multichannel return-to-
zero signal transmissions, IEEE Photon Technol. Lett. Vol. 8,
1996.
(b) [6] N. Shibata, K. Nosu, K. Iwashita, Y. Azuma,.Transmission
Fig. 4: (a) Eye Diagram at 100 µm Core Effective Area of Duobinary Limitations Due to Fiber Nonlinearities in Optical FDM
for 60 km Fiber Length, 3 ps/nm/km Dispersion and 0.8nm Channel Systems., IEEE Journal of Selected Areas in Commun., Vol.8,
Spacing. (b) Eye Diagram at 100 µm Core Effective Area of Binary No.6, 1999
for 60 km Fiber Length 3 ps/nm/km Dispersion and 0.8 nm [7] G.P. Agrawal, Nonlinear fiber optics, (Optics and Photonics, 4rd
Channel Spacing ed., Ac. Press, San Diego, 2008).
[8] R.-J. Essiambre and P.J. Winzer, “Fibre nonlinearities in
electronically predistorted transmission,” presented at the Eur.
IV. CONCLUSION Conf. Optical Commun. (ECOC’05), Glasgow, U.K., Sep. Paper
Tu 3.2.2, pp.25–29, 2005.
Try to Conclude That Which System is Better and [9] Yano, Y., et al. “2.6 terabit/s WDM transmission experiment
Why using optical duobinary coding”. Proc. European Conf. on
In nutshell for 3 ps/nm/km dispersion when fiber Optical Communication (ECOC), Paper ThB3.1, 1996.
[10] Bissessur, H., et al. “3.2 Tbit/s(80x40 Gbit/s) Phase-shaped
length is fixed that is 60 km duobinary is better than binary transmission over 3x100 km with 0.8 bit/s/Hz
binary format in terms of BER for 0.8 nm channel efficiency”,Electron.Lett. 38, pp. 377-379, 2002.
spacing when core effective area is varies from 2.5 to [11] Charlet, G., et al. “6.4 Tbit/s (159 x 42.7Gb/s.) capacity over 21
100 µm. For dispersion 3 ps/nm/km for fixed fiber x100 km using bandwidth-limited phase-shaped binary
transmission”. Proc. European Conf. on Optical Communication
length 60 km, when core effective area is varied from (ECOC), postdeadline paper PD4.1, 2002.
2.5 to 100 µm binary lags behind than duo binary in
terms of BER or duobinary show more improvement in
BER as compare to binary format and eye diagram also
clarifies same. It shows duo binary is more immune to
dispersion for variation in core effective area from 2.5
to 100µm.
Abstract—Apart from viruses, worms, Trojan horses bandwidth, storage and computational power. Using this
and network intrusions, there is a less familiar and very approach, bots are able to communicate with other bots
growing threat that tends to become harmful and by downloading files or commands from other bots’
disastorous botnets. A “botnet”consists of a network of machines and performing different activities. In
comproised computers controlled by an attacker
(“botmaster”). Nowadays, botnets have become the root
comparison to IRC structures, everyone can join a P2P
cause of many Internet attacks. Bots are controlled by a network, thus, the more peers acting as bots, the more
command and control (C&C) infrastructure to launch powerful the botmaster can be[2].Before the evolution
such attacks as Distributed-Denial-of-Service (DDoS), of Botnets; the major sources of malware were viruses,
spamming, identity theft and phishing, all of which worms, Trojan Horses that used to affect only a single
seriously threaten the Internet services and users. Prior machine. With the evolution of Botnets; the concept of
the botnets used usually organized themselves in a destruction was enhanced from a single machine to a
hierarchy architecture, which offers professionals network as a whole. The history of undertaking botnets
opportunities to detect or defend the botnets in their for destruction roughly dates back to 1990. Prior to
servers. However, newly-appeared P2P botnet, are
revealing a decentralized feature, which brought
this, botnets were the major sources of maintaining
difficulties in detection and mitigation.It simply improve control of the IRC channels. Their mischievous
the single point failure occur in centralized botnet.This applications mainly took advantage of the centralized
paper aims to provide a concise overview of botnets and control of IRC for command and control[3]. But
the detecting techniques. centralized control structure was relatively easy to
discover and track. Due to insecure nature of IRC
Keywords: Bot, Botnet, Botnet Detection, C&C, IRC.
botnets; they completely changed their structure form
centralized to a peer-to-peer nature, which is a
I. INTRODUCTION
decentralized control structure. This ultimately makes it
A “botnet” is a network of compromised machines much harde to spy the communication among the bots
(“bots”) connected to the Internet that is controlled by a and to track their origin. The command and control
remote attacker (“botmaster”) bots in the botnet channel is stable during the operation of botnets i.e.
connect directly to some special hosts (called once a botnet is established; the command and control
“command-and-control” servers, or “C&C” servers). channel remain the same throughout its operation. But
These C&C servers receive commands from the on the other hand, once a C&C channel is detected,
botmaster and forward them to the other bots in the then the whole botnet is exposed[4].
network. From now on, we will call a botnet with such
a control communication architecture a “C&C
botnet”[4]. From a botmaster’s perspective, the C&C
servers are the fundamental weak points in current
botnet architectures. First, a botmaster will lose
control of her botnet once the limited number of C&C
servers are shut down by defenders. Second, defenders
could easily obtain the identities. Third, an entire
botnet may be exposed once a C&C server in the
botnet is hijacked or captured by defenders[7].So after
that p2p botnet came into existence which has several
advantages over simple botnet. As a result, a new
approach of botnet structures starts to appear taking
advantage of existing Peer-to-Peer (P2P) protocols. The
attackers start to use P2P networks in order to control
their botnets. By using this approach, the bots can
contact other bots without having a centralised point for
their command and control (C&C) structure[1]. In P2P, Fig. 1: C&C Architecture of the P2P Botnet [4]
each node acts as client-server which provides
II. BACKGROUND AND MOTIVATION In operator mode, the user has elevated privileges that
allow them to control the room. In user mode, the user
Botnet Lifecycle Botnets share a common lifecycle
is simply a participant in the room and cannot control
that is dependent on the clients participating within the
room settings[8].
botnet. Each botclient will go through a similar
Typically, bots or zombies will enter a channel
process[3][1][7].
on an IRC server specified by the bot owner. The
• Exploitation: First, the vulnerable client is bot will have user privileges while a bot master, or
infected with bot software so that it can be controller, will have operator privileges. Bots will listen
controlled remotely. The exploit may be a for command messages sent across the channel by an
result of an unpatched vulnerability or operator[10].
malicious code that an unsuspecting user
executed on the local machine[3][1].
• Rallying: Typically, once the botclient is
infected, it will notify a botherder that it is
now a participant in the botnet. This is done
using the botnet owner s communications
method of choice. However, most botnets
currently use IRC channels for
communication[3].
• Security and Evasion: Once the machine has
been infected, bot software may try to take
measures to cover its tracks and remain an
active host. The botclient may try to
download additional modules that work to
prevent the user from knowing that their
machine is compromised or it could install
itself as a rootkit hidden deep within the
Operating System. To further prevent
detection, many botnets try to disable common
anti- virus software installed on the Fig. 2: Botnet Life Cycle[3]
machine[3].
• Receiving Commands: For the majority of its III. STATES-OF–ART
life, a botnet client will run quietly on a Jian kang et al. [2] proposed the basic principles
machine and wait for commands from a and mechanism of this decentralized P2P botnet, and
botherder or botnet owner. Botclients will present a novel detecting method using Multi-chart
typically listen for commands on an IRC CUSUM. propose a novel detecting method using the
channel. When a command is sent out across Multichart CUSUM test in detecting storm botnet.
the channel, the botclient will process and Kaulfman algorithm is also applied on the dynamic
execute it[3]. threshold adjusting to improve the detecting precision.
• Death: When a bot owner can no longer The results prove that the method can detect the storm
participate with one of its clients, it is no botnet in a relatively high precision with both low
longer considered a part of the botnet. This false- positive and false-negative. planning to work on
can happen for several reasons, including ways of mitigating and preventing those botnets in the
detection or removal of the malicious bot future. To detect this botnet, a Multi-chart CUSUM
software by the victim. If the botherder algorithm with its input on UDP, ICMP, and SMTP
suspects that their cover has been blown, they abnormally increasing is proposed. propose a method of
may take steps to cover their tracks by integrating several factors as our detecting characters,
executing a command that tells the bot testing them by improved change-point detecting
software to erase itself[3]. algorithm—Multi-chart CUSUM, and adjusting the
thresholds by Kaulfman algorithm. This prototype
A. Communication Techniques system effectively improves the precision in new P2P
The most common method for botnets to botnet detection and got relatively satisfied experiment
communicate is the Internet Relay Chat (IRC) protocol. results.
This protocol was designed for instant messaging using Sang-kyun et al. [10] proposed useful method for
client-server architecture. IRC servers host both public modeling multi-phased flows of P2P botnet traffic.
and private chat rooms known as channels. Participants Botnets are becoming more sophisticated and more
within a channel can have one of two modes of access. dangerous each day and attackers use the P2P protocol
to avoid centralized botnet topologies. We focus on Al-hamaadi et al. [1] proposed a correlation
the feature that a peer bot generates multiple traffic to algorithm to detect bots on the system by correlating
communicate with large number of remote peers. In their behavioural activities. results show that
this case, phased botnet flows have similar patterns, correlating different different activities can enhance
which occur at irregular intervals. We compress the detection mechanisms and reduce the false
duplicated flows via flow grouping and construct a alarms.Correlation algorithm is used because P2P
transition model of the clustered flows using a bots are difficult to detect as there is no central point of
probability-based matrix. A flow state is decided by communications. In addition, analysing network traffic
features consisting of; protocol, port, and traffic. Our looking for signatures can be tedious task because bots
model involves transition information about the state signatures can be dynamic and encrypted. different
values. Finally, we use the likelihood ratio for peer-to-peer bots such as Sinit and Nugache to examine
detection. In the experimental evaluation, we show the the behaviour of these bots. In their analysis, they note
efficiency of our proposed system with the that some peer-to-peer bots communicate on a fixed
SpamThru, Storm, and Nugache botnets Ping wang port. They argue that by monitoring traffic on that port,
et al. [5] proposed the design of an advanced hybrid one could detect these bots. They also discover that
peer-to-peer botnet. Compared with current botnets, the some of these bots generate a large number of
proposed botnet is harder to be shut down, monitored, destination unreachable error messages (DU) and
and hijacked. It provides robust network connectivity, connection reset error messages while trying to connect
individualized encryption and control traffic dispersion, to other peers. In addition, some bot’s communications
limited botnet exposure by each bot, and easy are encrypted which make the traffic analysis a difficult
monitoring and recovery by its botmaster. To defend task and resulting in high false alarms.
against such an advanced botnet, point out that Yuanyuan Zeng et al. [8] proposed a C&C
honeypots may play animportant role. Therefore, protocol- independent botnet detection framework that
invest more research into determining how to deploy combines both host-and network-level information. Our
honeypots efficiently and avoid their exposure to network flow analyzer searches for trigger action traffic
botnets and botmasters. Honeypot is an effective way to patterns among different hosts without accessing the
trap and spy on malware and malicious packets’ payloads, and clusters similarly behaving hosts
activities.Because compromised machines in a botnet into suspicious groups.Our host analyzer then obtains
need to cooperate and work together, it is suspicion-level information along with a few network
particularlyeffective to use honeypot techniques in statistics on a host-by-host basis for verification.
botnet spying. Finally, our correlation engine generates a detection
Wen-hwao et al. [7] proposed a P2P botnet result for each host by taking into account both
detection method relying on monitoring traffic at the suspicionlevel and clustering results. Our experimental
gateway and using data mining technology to analyze evaluation based on real-world data has shown the
network behavior. The research, besides proving the following results. The network analyzer can be
accuracy of the method by the results, also achieved the effective in forming suspicious clusters of aggressive
following objects: 1. Using Anomaly detection bots but may fail to separate benign hosts from bot-
regardless of the packets were encrypted or not; 2. infected hosts if the latter are stealthy at the network
Achieved higher distribution in practice without level. When the stealthy bots are present, it is the host
installing any software in computer or changing any analyzer that provides correct detection results by
network or routing architecture; generating distinguishing suspicion levels. By using
3. made pre-warning instead of post review combined host-and network-level information, our
possible only for the network behavior in P2P botnet framework is shown to be able to detect different types
without any attack (such as syn flood and port scan) as of botnets with low false-positive and false-negative
exploration parameters; 4. P2P botnet flow was found rates.
among various mixed flows in the same computer Saad et al. [9] proposed the characterization of
without being disturbed. On the other hand, the research network traffic behaviors to detect P2P botnet
was conducted only for LAN environment with more command and control (C&C) phase. The detection of
distributed structure from the internet. Therefore, it was botnet C&C phase is very important since it allows the
necessary to be distributed in ISPs if we want to stop detection of bots that spread under the radar through
P2P botnet massively and effectively, indicating that malicious email, websites, file sharing networks, ad hoc
the NAT technology would be generally used in wireless network and before these bots attack their
internet causing all the flows integrating as the same victims or achieve their goals. We discussed the main
IP, which eventually made the judgment of P2P botnet requirements of an online botnet detection framework
flow more difficult since the characters were extremely and investigated the power of different machine
diluted. learning (ML) techniques that are commonly used in
the literaturein addressing these requirements. Although
the performance of these techniques was promising, when one bot node communicate with lots of remote
none of these techniques can satisfy all the nodes. This directional Botnet traffic hassimilar
requirements of an online botnet detection framework. characteristics. Compress the traffic whose state is
This underscores the need to investigate new ML decided by agreement, port, communication volume
technique or an hybrid of existing ones that can satisfy through traffic group and traffic clustering structured
all the requirements of online botnet detection as by Probability matrix. The record raised a Botnet
outlined above. This will be the main purpose of our detection system based on irregular abnormal
future work. We also intend in our future work to study network independent from agreement structure of
the use of our proposed approach to detect different Botnet. The system adopts definition and
classes of botnets (beyond P2P botnets). characteristics of Botnet, showing similar C&C
Jinzhigang et al. [3] introduces Botnet detection communication mode and malicious behavior pattern in
methods, studies detection mechanism towards P2P the same Botnet[6].
Botnets based on user behavior, and proposes a new
case to identify Peer2Peer Botnet. To test and verify B. IRC Detection Techniques
function of the method we provide, a simple experiment A lot of techniques have been proposed for IRC
platform is designed and implemented. there are two Botnet detection. The basis of all these techniques is
major Botnet Detection Methods:Honeypot Technology founding of packets either at network layer or
and Traffic Analysis Technology.Honeypot is a trap application layer. the mechanism of detection is
host the defender used specially in the network to suggested on the network layer level. Here the
tempt attackers. All the programs running on it could hierarchy between routers and the IRC server is
be regarded as rogue program; while Traffic Analysis explored in bottom-up manner i.e. the tracking initiates
Technology is a method that the testing staff analyze the from the victim and follows the path of infecting
traffic of abnormal network by Network Analysis routers till the origin (bot-herder).he has proposed a
Software. So this paper put forward a new detection framework which sniffs the network traffic, filters it on
theory model—P2P Detection Technology based on the basis of application layer protocol, and then
combination of user sociality and Traffic Detection segregates them into either righteous or saboteur IRC
Analysis Technology traffic just by contemplating the IRC chat contents. The
separating foundation between a normal human and
IV. BOTNET DETECTION TECHNIQUES botnet conversation is that the human language is
In many cases, botnet infections and attacks are alternating while the Botnet conversation is repeating.
unavoidable. However, network users and presents a pipelined approach which accomplishes the
administrators must be aware of compromises and detection procedure in a number of steps. First it
attacks in order to remediate them. Several tools and separates the black and white list traffic based on the
techniques exist for detecting botnet infections when DNS queries; this separated traffic is classified
preventative measures have failed. There are also tools according to applications i.e. extract chat- like traffic.
that help network administrators detect when they ve Next pair wise correlation of the traffic flows is done to
become a victim of an attack[11]. identify similar traffic considering it to be originating
from same botnet. The study of these IRC detection
A. Two Major Botnet Detection Methods techniques reveals that choice of the suitable detection
technique depends on the required scenario. If the
Honeypot Technology and Traffic Analysis solution has to be managed at the network layer, serves
Technology. Honeypot is a trap host the defender as the best option; while on application layer ans serve
used specially in the network to tempt attackers. All the purpose[5].
the programs running on it could be regarded as Again we categorized botnet detection strategies
rogue program; while Traffic Analysis Technology is
broadly into 2 types:
a method that the testing staff analyze the traffic of
abnormal network by Network Analysis Software. But 1) Host based detection
the premise of Honeypot Technology is that the
honeypot must be added in Botnet concealedly and Host based detection pertains to detecting bot
ensure that it will not be antidetected by attackers. activities on a single machine. Some typical symptoms
Most Botnet of P2P Agreement hold a tight through which botnets can be detected via host based
authentication mechanism, making it hard to add detection are:
honeybot into botnet. In addition, most well-developed • Infection detection by antivirus. This may or
Traffic Analysis Technology is used to detect botnet may not be a botnet activity but certainly can
based on IRC Agreement, not suit the new botnet of of be a starting point for infection. Many
P2P Agreement yet. He has proposed a multi phase infections might not even be detected.
traffic model which was based on p2p botne • Slowing of the machine. Again, this can
traffic.which has focuses on the multi traffic produced happen due to variety of reasons but if this is a
sudden change, one must check for Low Interaction: Honeypots that limit what an
spyware/adware on the system using some attacker can do with a system are considered low
scanner[4]. interaction. The host or network services within the
o Detection of rookits on the machine. honeypot are normally emulated so that no device is
o Modification of Windows host / system actually infected. This method is primarily used for
files. malware collection[7].
o Random popups indicating adware High Interaction: When actual services and
presence on the machine which can also software are used and system compromises are not
be a form of botnet click fraud activity. simulated the honeypot is considered high interaction.
o If your DNS resolution server is not your In this case, a system is closely monitored while an
ISP's or company's server, then it might attacker compromises and infects the system. These
have been replaced by a shady source and honeypots allow security researchers to get a better
can forward your requests to shady understanding of how attackers and malware are
URLs. modifying a system in order to use it to their advantage.
It can also help researches discover new, stealthier
2) Network based detection exploits (Honeypots)[4][7].
Network based detection pertains to detecting bot
activities on a network. Some typical symptoms through V. BOTNET USES AND STRATEGIES
which botnets can be detected via network based Infecting machines across the world with control
detection are: mechanisms would be meaningless without some kind of
• One can sniff IRC traffic across commonly incentive for the attacker. Unfortunately, bots are used
used IRC (Internet Relay Chat) ports. for network attacks as well as making profit. Recently,
• Most common ports used for IRC is port 6667. more botnets are being designed to help their owners
Many bot masters today use non standard IRC make profit instead of simply denying services[9].
ports for communication to avoid detection.
So, observing the suspicious outbound A. Denial of Service
connections on non standard ports would be a One of the first uses of botnets was in Denial of
good idea for detecting bot activities. Also, Service (DoS) attacks. These attacks seek to overload a
the IRC traffic being generally in plain text, target, such as a website or server, until it is completely
sniffing for known IRC commands and unresponsive to legitimate traffic. Specifically, botnets
keywords will help detecting a botnet most often perform a DoS attack by commanding
activity[4][6]. zombies from all over the internet to send huge
• By maintaining a check-list of known C&C amounts of network traffic to a target. This is known as
servers, one can check for outbound requests to a Distributed Denial of Server (DDoS) attack.
those servers. Attackers use several techniques to perform these
• If many machines on the network are attacks[9][1].
accessing same server at once, bot masters
keep changing their DNS servers to shift their B. TCP SYN Flood
location. In a TCP SYN Flood attack, the attacker
• Keeping a check on ports 135, 139, 445 commands zombies to send SYN messages, or the
(ports for windows file sharing) may also first part of the TCP handshake, without responding to
help detect presence of bots. A heavy traffic follow-up requests. This attack may also be combined
over these ports, is an indication of some bot with IP spoofing so that follow-up requests are sent to
activity. another host. As thousands of SYN messages are sent to
• Huge amount of SMTP out bound traffic, the server, the target s connection table fills up and
especially from servers which are not large amounts of memory must be allocated to each
supposed to be SMTP servers indicates connection attempt until it exceeds the timeout period.
infection of malware spam in the network. Eventually, the server may not be able to handle any
One of the most popular methods for tracking and more SYN requests and therefore may start dropping
analyzing botnets is the honeypot[4]. A honeypot is a traffic[9].
network host set up by security researchers and
disguised as a vulnerable machine that attackers might C. UDP Flood
target. Once the honeypot has been compromised by an Using the UDP protocol, an attacker may attempt
attacker, security researchers can analyze the bot in to overwhelm a server by having botnet zombies
order to get information about the attack and its source simultaneously sent small UDP packets to the target.
(Honeypots). Honeypots can generally be grouped into These packets are sometimes sent on random open
two different categories:
ports. Once the packets reach the target, it must allocate complexity of botnets continues to increase. New
resources to handle them. If enough packets reach the methods for intrustion, infection, and evasion are
target, it can cause other traffic to be dropped[7]. making it increasingly difficult to track and detect bots.
There are several ways that individuals and groups can
D. Smurfing lessen the threat of botnets. Keeping systems up-to-
Smurf attacks involve sending an overwhelming date, installing firewalls, and making users aware of
number of ICMP ping requests to a broadcast address common social engineering strategies are a few of the
while spoofing the source IP address. The spoofed ways individuals as well as network administrators can
source IP address is often the target of the attack prevent their computers from becoming part of a
since each host in the broadcast range will attempt to botnet. This could suffocate botnets by eliminating
reply to the source address that was specified. This can the resources they need to become massive
overwhelm the target with requests. However, there
are newer techniques to filter this spoofed, or REFERENCES
illegitimate, traffic at the border of the network, [1] Y. Al-Hammadi and U. Aickelin, "Behavioural Correlation for
preventing this kind of attack from being effective[3]. Detecting P2P Bots," in Future Networks, 2010. ICFN '10.
Second International Conference on, pp. 323-327.
1) Click fraud attack [2] K. Jian, Z. Jun-Yao, L. Qiang, and L. Zhuo, "Detecting New
P2P Botnet with Multi-chart CUSUM," in Networks Security,
Online advertising programs that offer site owners Wireless Communications and Trusted Computing, 2009.
compensation for advertisement clicks have also fallen NSWCTC '09. International Conference on, 2009, pp. 688-691.
[3] Z. Jin, Y. Wang, and B. Wei, "P2P Botnets detection based
victim to botnet attacks. By programmatically on user behavior sociality and traffic entropy function," in
commanding infected machines to click on an Consumer Electronics, Communications and Networks
advertisement on a web page, an attacker can generate (CECNet), 2012 2nd International Conference on, pp. 1953-
thousands of unique, but illegitimate, clicks instantly for 1955.
[4] Z. Ling Yun, "VMM-Based Framework for P2P Botnets
a substantial profit Tracking and Detection," in Information Technology and
Computer Science, 2009. ITCS 2009. International Conference
2) Keylogging on, 2009, pp. 172-175.
[5] W. Ping, S. Sparks, and C. C. Zou, "An Advanced Hybrid Peer-
Attack Some botnets are organized to log to- Peer Botnet," Dependable and Secure Computing, IEEE
keystrokes on each infected machine in order to steal Transactions on, vol. 7, pp. 113-127.
sensitive information from users. In some cases, some [6] Wang, T. Li, and H.-b. Wang, "Botnet Detection Based on
bots can also filter logged keystrokes for keywords that Analysis of Mail Flow," in Biomedical Engineering and
Informatics, 2009. BMEI '09. 2nd International Conference
may indicate a username and password sequence has on, 2009, pp. 1-4.
been entered. For example, the keyword “Ebay” could [7] L. Wen-Hwa and C. Chia-Ching, "Peer to Peer Botnet
indicate that someone is typing the address to Ebay and Detection Using Data Mining Scheme," in Internet Technology
is about to sign in with their credentials. This can be and Applications, 2010 International Conference on, pp. 1-4.
[8] Z. Yuanyuan, H. Xin, and K. G. Shin, "Detection of botnets
more effective than sniffing network traffic for user using combined host- and network-level information," in
credentials since it bypasses any encryption that web Dependable Systems and Networks (DSN), 2010 IEEE/IFIP
sites or services are using. International Conference on, pp. 291-300.
[9] S. Saad, I. Traore, A. Ghorbani, B. Sayed, D. Zhao, L. Wei,
J. Felix, and P. Hakimian, "Detecting P2P botnets through
VI. CONCLUSION network behavior analysis and machine learning," in Privacy,
Security and Trust (PST), 2011 Ninth Annual International
Use of botnet has to led a rise in DDOS attacks on Conference on, pp. 174-180.
resulting in significant loses to the economy.These [10] N. Sang-Kyun, O. Joo-Hyung, L. Jae-Seo, N. Bong-Nam, and
network can be harmfull for constructive J. Hyun-Cheol, "Detecting P2P Botnets Using a Multi-phased
purpose,however currently they are primarily being Flow Model," in Digital Society, 2009. ICDS '09. Third
International Conference on, 2009, pp. 247-253.
used to penetrate computer related crime such as [11] K. Jian and Z. Jun-Yao, "Application Entropy Theory to
SPAM&DDOS. Response to a botnet attack is Detect New Peer-to-Peer Botnet with Multi-chart CUSUM," in
difficult.However,if proper precaution are undertaken Electronic Commerce and Security, 2009. ISECS '09. Second
prevention is possible. The impact that botnets have International Symposium on, 2009, pp. 470-474.
[12] Z. Yuanyuan, Y. Guanhua, E. Stephan, and G. S. Kang,
had on networks and the internet since their inception "Measuring the effectiveness of infrastructure-level detection of
really demonstrates one of the greatest threats to large-scale botnets," in Proceedings of the Nineteenth
computer systems in a digital age. If previous and International Workshop on Quality of Service San Jose,
current botnet strategies are any indication, the problem California: IEEE Press.
is likely to get worse before it gets better. The
III. ATTACKS POSSIBLE IN AD-HOC NETWORK 9. Black hole attack: Like original Black hole in
universe, this node absorbs anything (any
In active attack, attacker can modify packets,
packet) whatever is going through it. In other
absorb packets, inject packets, or even disrupt network
words it drops all the packets rather than
services. However in case of passive attack, an attacker
forwarding them to its original path or
passively read the packets that are passing trough it
location. Since a black hole does not check its
without modifying or tampering packets.
routing table so it can reply fast as compare to
Main attacks are:
original node [5].
1. Route Logic Compromise: Instead of original
10. Gray hole attack: Unlike black hole attack it
routing control message incorrect routing
drops only selected packets
control messages are injected to damage
11. Wormhole attack: In this type of attack,
routing logic.
attacker requires to have two nodes, connected
2. Traffic Distortion Attack: All attacks that
by private tunnel. Two nodes can connected
create problem for data packets to transfer
through a wired or wireless link. The
from the source to the destination. This type of
tunnelling will be always faster than the
attack can snoop network traffic, corrupt
normal multi-hop route (source to destination).
packet, block or reply transmissions for some
This is a repeater attack; an attacker records
malicious purposes.
the packets at one place and tunnel to another
3. Jamming: Wanted and unwanted interference
location and replies from the second location.
can happen with radio waves of MANETs,
This is a very powerful attack because it can be
because WLANs use unlicensed frequencies.
done if all communication provides
Other electromagnetic devices can operate on
authenticity or digital signature and even if
same that can overlap with WLAN traffic. If
attacker has no keys.
attacker has a powerful transmitter and can
12. Eavesdropping: In this, packet can be decoded
generate a strong radio signal that can corrupt
during the journey and forwarded to next as it
a weak signal, this condition is called
is.
jamming. Jammers can be of two types:
13. Steep deprivation: In this attack a node tries to
• High power pulse full band jammers. save its own energy.
• Low power partial-band jammers 14. Sybil attack: A node acquires more than one
The solution to jamming is to use Spread-Spectrum identity.
technology to transmit data. 15. Selfishness: A node transmits and receives
4. Spoofing: source address can be corrupt by only own packets however that does not
injecting data and control packets. behave a part of network.
5. Flooding attack: one or many node sends 16. Routing disruption attacks: In MANET,
floods of Route Request (RREQ) and Route sometimes an attacker can generates routing
Reply (RREP) messages meaninglessly. This control packets and they are generally a
is called flooding attack. Solution for this falsified routing REPLY ( in case of dynamic
attack is to authenticate route control messages source routing protocol. Such type of control
[4]. packets prevents to find out correct path and
6. Packet Modifications and Dropping: correct routing logic in the whole network.
Intermediate nodes can modify the packet
content. Also it is possible to change the IV. HISTORY OF INTRUSION DETECTION SCHEME
header information including source-
destination addresses. The malicious In 1980, James Anderson wrote report for US air
intermediate nodes can also drops data. force to propose a unique method to filter out audit
7. Repeater attack: Two nodes are said to be trails and detect unusual attacks. In 1986, Dorothy
neighbour if they are in transmission range Denning and Peter developed a real time intrusion
directly to each other. Any attacker node A detection system. In 1990 US Davis developed another
shows, that node A is neighbour of node B and system i.e., Network system monitor. In 1992,
the service of node B is prevented from Distributed intrusion detection system developed. In
receiving and sending data packets to its 1998 DARPA sponsored an intrusion detection
destinations. This type of attack can be evaluation of many IDS. In 2000, this becomes an open
protected by Neighbour Detection Techniques. research topic for whole world. [2]
8. Identity impersonation: Generally there is no In 2002, Oleg Kachirski and Ratan Guha have
authentication scheme take place in MANET. proposed a distributed multisensory Intrusion detection
Therefore the IP address and MAC based scheme. After that, Yian Huang, Wenke Lee proposed a
identity are easy to impersonate, if cooperative Intrusion detection scheme for ad-hoc
communication channel is not secured. network. In this, there are six modules which work
same as previous but if proof against an attacker node is others the malicious action of the node. It is a type of
not sufficient then a request is sent to global response detecting attacks inside and second one is Network
module for cooperating. Further Ricardo Puttini et al based intrusion detection. It captures and examines the
developed an algorithm which was fully distributed. In network traffic that flows through it. There is no central
this, data has been collected but execute algorithm and body that monitor to whole network therefore this
alerts to neighbour nodes. Further in 2005, Abdul scheme is not suitable for each case.
Rehman Hijazi and Naseer proposed a scheme by which There are three components of Data analysis. First
bandwidth can be conserved by using local stationary is IDS based on misuse detection. Misuse detection IDS
and mobile agent. Satoshi Kurosawa proposed a model analyze the collected data for well known attacks
of IDS based on AODV in November 2005. And further patterns, sometimes these patterns are known as
in the same year Geethapriya Thamilarasuet.al signature (Set of patterns of known attacks). Alarm
presented a new model of IDS based on cross layer. whenever signature matches. The main drawback of
Cao Minh Trang et al implemented a model based on misuse detection is that it can only detect previous well
on-demand Distance vector protocol. Yang fang Fu et al known attacks. Another type of Intrusion detection is
proposed hierarchical and distributed IDS to find out based on anomaly (behavior based). Anomaly detection
the status of every node in network says cluster. Inthis scheme check the abnormal behavior of any node. Any
scheme there are two types of Intrusion detection first type of behavior that does not match a normal behavior
one is host based and other one is network based. that comes in the category of attack. In this technique, a
Jaydeep sen et al proposed a reputation system for threshold is set so that the difference can be created in
MANET. In reputation system incentives are provided. between a normal node and an attacker node behavior
This system is based on six modules (i) monitoring (ii) patterns. The main advantage of this scheme is that it
reputation collection (iii) reputation maintenance (iv) can detect a new type of unknown attack. Third type of
reputation formatting (v) reputation propagation (vi) IDS is based on specification. In this, IDS defines a set
alarm raiser module.In 2010, Abolfazl Esfandi proposed of rules that explains the correct operation or set of
a scheme in which data has been collected using data rules and this protocol monitors the overall process.
mining technique. This gives many advantages that it Narrower or incomplete signatures (false negative) will
increases alarm rate, detection rate, false positive is miss some attacks sometimes. A broader signature
reduced. In 2011 Ghenima Bourkach et al proposed a (false positive) will raise false alarms. It is necessary to
distributed model made of local IDS agent which is update protocol for new known attack patterns. Each
same as earlier in many respects. [7] mobile node runs IDS independently. Each mobile host
believe on its own observation table and analysis but
V. PROBLEMS OF WIRED IDS TECHNIQUES cooperation is possible.
There are various problems comes but some main
are given below.
• Real time traffic analysis is no longer possible
in adhoc network [6].
• Ad hoc network does not have gateway and
routers therefore whole network is to be
considered for audit data [6].
• Audit trace is limited in ad hoc network [6].
• Ad-hoc nodes may be stingy about
communication because slower links, limited
bandwidth, higher cost, and battery power
scarcity [6].
REFERENCES
[1] Oleg Kachiiri, Ratan guha “intrusion detection using Mobile
agents in Wireless adhoc networks” Proceeding of IEEE
workshop on Knowledge Media networking IEEE 2002, pp:153-
158.
Fig. (d): Functional Diagram of IDS [2] www.engr.smu.edu/tchen
[3] H. Debar, M. Dacier, and A. Wespi, “A Revised Taxonomy for [7] Rupali Pathak, Durgesh Kumar Mishra “ distributed intrusion
Intrusion-Detection Systems,” Annales des Tlcommunications, detection scheme for wireless adhoc networks: a review” CSI
vol. 55, 2000 6thInternational conference on software engineering (conseg-
[4] Ning, P.; Sun, K.: How to Misuse AODV: a Case Study of 2012) pp:358-363
Insider Attacks Against Mobile Ad-hoc Routing Protocols. In: [8] Y. Huang, and W. Lee, “A Cooperative Intrusion Detection
Proceedings of the 2003 IEEE Workshop on Information System for Ad Hoc Networks,” Proceedings of the ACM
Assurance, pp.60-67, NY (2003). Workshop on Security of Ad Hoc and Sensor Networks
[5] Shurman M.AI, Yoo S.M., Park S.: Black Hole Attack in (SASN’03), Fairfax VA, October 2003.
Wireless Ad Hoc Networks. In: Proceedings of ACM 42nd
Southeast Conference (ACMSE 04),pp. 96-97, Alabama (2004).
[6] Y. Zhang and W. Lee, “Intrusion Detection in Wireless Ad Hoc
Networks,” the 6th Annual International Conf. on Mobile
Computing and Networking (ACM MobiCom’00), Boston, MA,
Aug., 2000, pp. 275-283.
Abstract—In Wireless Sensor Netw work, sensor nodes Protocol Independent M Multicast-Spaarse Mode
or motes coommunicate with
w each otherr and also with h Base (PIM
M-SM) routes multicast pacckets to multiccast groups,
Station usiing their wireless radios frequency
f chaannel, and is designed too efficiently esstablish distrib
bution trees
allowing th
hem to dissemiinate their sennsor data to reemote across wide area networks (WA ANs). PIM-SM is called
processing, visualization, analysis, annd storage sysstems.
Wireless Seensor Network k is an emergiing technologyy that
Protoocol Indepenndent becausee it can usee the route
shows great promise for various futuristic applications. In inforrmation that any routing protocol enteers into the
this paper we
w mainly ideentify the archiitectural issues that multticast Routingg Information Base (RIB), or, as it is
are arise with the multicaasting protocolls in wireless sensor
s knowwn in Windoows terminoloogy, the multticast view.
networks. Exammples of these routing prrotocols inclu ude unicast
proto
ocols such as a the Routinng Informatio on Protocol
Keywords: Wireless Sensor nodes,, Network, Prootocol
Independennt Multicastingg, Dense Moode, Sparse Mode,
M (RIP
P) and Openn Shortest P Path First (O OSPF), but
Shortest Path Tree, Sharedd Tree multticast protocolls that populatte the routing tables such
as the
t Distance Vector Mullticast Routin ng Protocol
I. INTTRODUCTION (DVVMRP) can allso be used. SSparse mode means that
the protocol
p is deesigned for situations wherre multicast
A wirreless sensor node or motte has not onnly a grou
ups are thinlyy populated across a larrge region.
sensing coomponent, but b also on-bboard processsing, Sparrse-mode prrotocols cann operate in LAN
communicaation, and sttorage capabiilities. With these envirronments, butt they are mosst efficient oveer WANs.
enhancemeents, a senssor node is often not only
responsiblee for data colllection, but also
a for in-nettwork
analysis, correlation, annd fusion of itts own sensorr data
and data frrom other senssor nodes especially sink. WhenW
many sensors cooperaatively monitor large phyysical
environments, they forrm a wireless sensor netw work.
There exiists a varietty of sensoors that calcculate
environmental parameteers such as temperature, light
intensity, sound, magnnetic fields, image, etc. The
informationn based on sensed data caan also be ussed in
agriculturee and livestocck, assisted driving
d or even in
providing security
s at homme or in publiic places. To fulfill
f
privacy annd security requirements
r in an approppriate
architecturre for WSNs and offering pervasive serrvices
is essentiall for user acceeptance.
ISBN: 978-9
93-82880-20-2 245
Proceedingss of National Co
onference on Tre
ends in Signal Processing
P & Co
ommunication (TSPC’13) 12th–
–14th April 2013
3
The problem
p with flooding is that it createes an
exponentiaal number of copies
c of eachh packet. So, on
o the
one hand, flooding guuarantees thatt a copy of each
packet willl be delivereed to each noode, providedd that
packets arren't lost; on the other hand, floodingg can
generate soo much conggestion that packets
p very likely
l
will be lostt.
Fig. 1.5: Prune Messages in WSN
246 Bhag
gwant Institute of
o Technology, Muzaffarnagar,
M Uttar Pradesh, India
Protocol Independent Multicast—Sparse and Dense Modes in Wireless Sensor Network
The shared tree is unidirectional, which means data specific unicast routing protocol. Because it is a router-
flows only from the RP to the receivers. For a host other to-router protocol, all routers in the network must be
than the RP to send on the tree, the data must first be upgraded to support PIM-SIM higher version.
tunneled to the RP before it can be multicast to the
participants. This means that if a receiver is also a REFERENCES
source, it can't use that tree to send packets to the RP. It [1] A. Ballardie. Core Based Trees (CBT)Multicast Routing
can only use it to receive packets from the RP. Architecture. RFC 2201, September 1997.
RPTs have longer delays, packets must first be sent [2] Bommaiah, McAuley, and Talpade. AMRoute: Adhoc Multicast
to the RP before they can be distributed, but have less Routing Protocol. Internet-Draft, draft-talpade-manet-amroute-
00.txt, February 1999. Work in progress.
router state to maintain. Some examples of applications [3] Josh Broch, David A. Maltz, David B. Johnson, Yih-Chun Hu,
where an RPT is appropriate include: 1. Networks with and Jorjeta G. Jetcheva. A Performance Comparison ofMulti-
many low-rate data sources, 2. Applications that can HopWireless Ad Hoc Network Routing Protocols. In
tolerate delay, 3. Applications requiring consistent Proceedings of the Fourth Annual ACM/IEEE International
Conference on Mobile Computing and Networking, pages 85–
policy and access control across most participants in a 97, October 1998.
group, 4. Networks where most of the source trees [4] C.-C. Chiang, Mario Gerla, and Lixia Zhang. Forwarding Group
overlap topologically with the shared tree. Multicast Protocol (FGMP) for Multihop, Mobile Wireless
Conferencing applications can use both the SPTs Networks. ACM Baltzer Journal of Cluster Computing: Special
Issue on Mobile Computing, 1(2):187–196, 1998.
and the RPT. The RPT could be used for keep-alive [5] M. Scott Corson and Anthony Ephremides. A Distributed
packets, because these are sent at a low data rate. The Routing Algorithm for MobileWireless Networks. Wireless
sources could use the SPTs because they are sending Networks, 1(1):61– 81, feb 1995.
out a great deal of data. [6] Steve Deering. Host Extensions for IPMulticasting. RFC 1112,
August 1989.
Although each path from the RP to a receiver is the [7] Kevin Fall andKannanVaradhan, editors. ns Notes and
shortest path, the shortest path from the source to the Documentation. The VINT Project, UC Berkeley, LBL,
receivers is not the same as the path from the receiver to USC/ISI, and Xerox PARC, November 1997. Available from
the RP. A legitimate question is: Why not use the SPT http://www-mash.cs.berkeley.edu/ns/.
[8] J.J. Garcia-Luna-Aceves and E.L. Madruga. A Multicast
rather than the RPT? For PIM-SM, there are two Routing Protocol forAd-HocNetworks. In Proceedings of the
answers. First, PIM-SM has a method that allows the IEEE Conference on Computer Communications, INFOCOM
last-hop router to leave the RPT and join the SPT if the 99, pages 784–792, March 1999.
volume of traffic warrants it. This is called an SPT [9] Zygmunt J. Haas. A Routing Protocol for the Reconfigurable
Wireless Network. In 1997 IEEE 6th International Conference
switch, and is discussed later in the paper. Second, on Universal Person Communications Record. Bridging the
when the RPT is used, routers don't need to maintain as Way to the 21st Century, ICUPC ’97, volume 2, pages 562–566,
much state, which decreases the amount of memory October 1997.
required. [10] Hugh Holbrook and Brad Cain. Source-Specific Multicast for
IP. Internet-Draft, draft-holbrook-ssm-arch-01.txt, November
2000. Work in progress.
V. PROBLEM ISSUES ASSOCIATED WITH MULTICASTING [11] IEEE Computer Society LAN MAN Standards Committee.
ROUTING PROTOCOL Wireless LAN Medium Access Control (MAC) and Physical
Layer (PHY) Specifications, IEEE Std 802.11-1997. The
PIM-SM higher version is currently considered the Institute of Electrical and Electronics Engineers, New York,
de facto standard multicast routing protocol. However, New York, 1997.
some issues are still not resolved. As it is a router-to- [12] Jorjeta G. Jetcheva, Yih-Chun Hu, David A. Maltz, and David
B. Johnson. A Simple Protocol for Multicast and Broadcast in
router protocol, all routers in the network must be Mobile Ad Hoc Networks. Internet-Draft, draft-ietf-manet-
upgraded to support it. A second issue is that, because simple-mbcast-01.txt, July 2001. Work in progress.
the number of candidate RPs scales linearly with the [13] L. Ji and M. S. Corson. A Lightweight Adaptive Multicast
size of the domain, the protocol cannot scale globally. Algorithm. In Proceedings of IEEE GLOBECOM ’98, pages
Mapping G to RP may also present a scaling problem 1036–1042, December 1998.
[14] L. Ji and M. S. Corson. Differential Destination Multicast
because it involves flooding the BSR announcements. (DDM) Specification. Internet-Draft, draft-ietf-manet-ddm-
Another problem is the location of the RP. There are 00.txt, July 2000. Work in progress.
many ISPs in the world, and none of them want to [15] David B. Johnson. Routing in Ad Hoc Networks of Mobile
depend on an RP in the domain of another ISP for Hosts. In Proceedings of the IEEE Workshop on Mobile
multicast service between its own customers. Computing Systems and Applications, pages 158–163,
December 1994.
[16] David B. Johnson and David A. Maltz. Dynamic Source
VI. CONCLUSION Routing in Ad Hoc Wireless Networks. In Mobile Computing,
edited by Tomasz Imielinski and Hank Korth, chapter 5, pages
PIM-SM is the de facto standard multicast routing 153–181. Kluwer Academic Publishers, 1996.
protocol. It is designed to perform efficiently in WANs, [17] David B. Johnson, David A. Maltz, Yih-Chun Hu, and Jorjeta G.
where multicast groups are sparsely distributed. It Jetcheva. The Dynamic Source Routing Protocol for Mobile Ad
Hoc Networks. Internet-Draft, draft-ietf-manet-dsr-05.txt,
maintains the traditional IP multicast service model of March 2001. Work in progress.
receiver initiated membership and supports both shared [18] S.Kent andR.Atkinson. Security Architecture for the Internet
and shortest path trees. PIM-SM is not dependent on a Protocol. RFC 2401, November 1998.
[19] S.-J. Lee, Mario Gerla, and C.-C. Chiang. On-Demand Multicast [23] P. Sinha, R. Sivakumar, and V. Bharghavan. MCEDAR:
Routing Protocol. In Proceedings of the IEEE Wireless Multicast Core Extraction Distributed Ad-Hoc Routing. In In
Communications and Networking Conference, WCNC ’99, Proceedings of the Wireless Communications and Networking
pages 1298–1304, September 1999. Conference, WCNC ’99., pages 1313–1317, September 1999.
[20] S.-J. Lee, W. Su, J. Hsu, M. Gerla, and R. Bagrodia. A [24] C.-K. Toh, Guillermo Guichala, and Santithorn Bunchua.
Performance Comparison Study of Ad Hoc Wireless Multicast ABAM: On- Demand Associativity-Based Multicast Routing for
Protocols. In Proceedings of IEEE INFOCOM 2000, pages 565– Ad Hoc Mobile Networks. In Proceedings of IEEE Vehicular
574, March 2000. Technology Conference, VTC 2000, pages 987–993, September
[21] The Monarch Project. Rice Monarch Project: Mobile 2000.
Networking Architectures, project home page. Available at [25] C.W.Wu, Y.C. Tay, and C-K. Toh. Ad hoc Multicast Routing
http://www.monarch.cs.rice.edu/. protocol utilizing Increasing id-numberS (AMRIS). Internet-
[22] Elizabeth M. Royer and Charles E. Perkins. Multicast Operation Draft, draft-ietfmanet-amris-spec-00.txt, November 1998. Work
of the Ad-hoc On-Demand Distance Vector Routing Protocol. In in progress.
Proceedings of the Fifth Annual ACM/IEEE International
Conference on Mobile Computing and Networking, Mobicom
’99, pages 207–218, August 1999.
expensive or dangerous, which is another performance. Usually emulator has highly scalability,
reason for using simulation [4]. which can emulate numerous sensor nodes at the same
• Additionally, knowing the weaknesses of time.
available simulators could help developers to
identify drawbacks of their own models, when C. Type of Network Simulators
compared with these simulators, thus Different types of network simulators can be
providing an opportunity for improvement. categorized and explained based on some criteria such
In this paper we conduct a survey of several as if they are commercial or free, or if they are simple
simulators that we find significant and interesting. In ones or complex ones or if they are Trace- Driven
the second section, we present basic concepts in Simulation and Discrete-Event Simulations [Jain91].
network simulation. In the third section, we describe
methodology and criteria. In the fourth section, we 1) Commercial and open source simulators
describe the different simulators. In the fifth section, Some of the network simulators are commercial
simulators are compared and analyzed according to which means that they would not provide the source
methodology and criteria and in the sixth section, we code of its software or the affiliated packages to the
conclude the paper. general users for free. All the user s have to pay to
get the license to use their software or pay to order
II. BASIC CONCEPTS IN NETWORK SIMULATION specific packages for their own specific usage
In this section, we will introduce some basic requirements. The advantage is that it generally has
concepts in the area of network simulation complete and up-to-date documentations and they can
be consistently maintained by some specialized staff in
A. Simulation and Emulation that company.
In the communications networks, simulation is a While the open source network simulator is
useful technique since the behavior of a network can disadvantageous in this aspect, and generally there
be modeled by calculating the interaction between the are not enough specialized people working on the
different network components (routers, physical links documentation. On the contrary, the open source
or packets) using mathematical formulas. Another network simulator has the advantage that everything is
characteristic of network simulation is that the very open and everyone or organization can contribute
simulation program can be used together with different to it and find bugs in it. The interface is also open for
applications and services in order to observe end-to-end future improvement. It can also be very flexible and
or other point-to-point performance in the networks. reflect the most new recent developments of new
Network emulation means that network under technologies in a faster way than commercial network
planning is simulated in order to assess its performance simulators.
or to predict the impact of possible changes, or
TABLE 1 NETWORK SIMULATORS
optimizations. The major difference lying between them
is that a network emulator means that end-systems such Network Simulators Name
as computers can be attached to the emulator and will Commercial OPNET, QualNet
Open Source NS2, NS3, OMNeT++, SSFNet, J-Sim
act exactly as they are attached to a real network. The
main point is that the network emulator's job is to 2) Simple vs. complex
emulate the network which connects end-hosts, but not
the end-hosts themselves. Currently there are a great variety of network
simulators, ranging from the simple ones to the
B. Simulator and Emulator complex ones. Minimally, a network simulator should
Simulator [8] is universally used to develop and enable users to represent a network topology, defining
test protocols of WSNs, especially in the beginning the scenarios, specifying the nodes on the network, the
stage of these designs. The cost of simulating thousands links between those nodes and the traffic between the
of nodes networks is very low, and the simulation can nodes. More complicated systems may allow the user to
be finished within very short execution time. Both specify everything about the protocols used to process
general and specialized simulators are available for uses network traffic. Graphical applications also allow users
to simulate WSNs. to easily visualize the workings of their simulated
The tool, which is using firmware as well as environment. Some of them may be text- based and can
hardware to perform the simulation, is called emulator provide a less visual or intuitive interface. Others may
[8]. Emulation can combine both software and be programming-oriented and can provide a
hardware implementation. Emulator implements in programming framework that allows the users to
real nodes, thus it may provide more precision
customize to create an application that simulates the and limitation. This section illustrates five main- stream
networking environment for testing. simulation tools used in WSNs: NS-2, TOSSIM,
OMNeT++, J-Sim, Avrora, and analyzes the merits
3) Discrete-Event simulations and trace-driven and limitation of each simulation tool.
simulation
A. NS-2
Discrete-event simulation [9, 10] is widely used in
WSNs, because it can easily simulate lots of jobs Overview: Ns-2[2,6,7,11,12,13,18,19,20] is the
running on different sensor nodes. Discrete- event most popular simulation tool for sensor networks. It
simulation includes some of components. This began as ns (Network Simulator) in 1989 with the
simulation can list pending events, which can be purpose of general network simulation. Ns-2 is an
simulated by routines. This simulation includes input object- oriented discrete event simulator; its modular
routines, output routines, initial routines, and trace approach has effectively made it extensible.
routines. In addition, this simulation provides dynamic Simulations are based on a combination of C++ and
memory management, which can add new entities and OTcl. In general, C++ is used for implementing
drop old entities in the model. Debugger breakpoints protocols and extending the ns-2 library. OTcl is used to
are provided in discrete-event simulation, thus users can create and control the simulation environment itself,
check the code step by step without disrupting the including the selection of output data. Simulation is run
program operation. at the packet level, allowing for detailed results.
However, Trace-Driven Simulation [9] provides Merits: NS-2 can support a considerable range
different services. This kind of simulation is commonly of protocols in all layers. For example, the ad-hoc and
used in real system. The simulation results have more WSN specific protocols are provided by NS-2.
credibility. It provides more accurate workload; these Secondly, the open source model saves the cost of
detail information allow users to deeply study the simulation, and online documents allow the users easily
simulation model. Usually, input values in this to modify and improve the codes.
simulation constant unchanged. However, this Limitation: People who want to use this simulator
simulation also contains some drawbacks. For example, need to familiar with writing scripting language and
the high-level detail information increases the modeling technique; the Tool Command Language is
complexity of the simulation; workloads may change, somewhat difficulty to understand and write. NS-2 is
and thus the representativeness of the simulation needs more complex and time-consuming than other
to be suspicious. simulators to model a desired job. NS-2 provides a
poor graphical support, no Graphical User Interface
III. METHODOLOGY AND CRITERIA (GUI).
We define a set of criteria to compare the B. OMNeT++
selected simulators. The criteria are:
Overview: OMNeT++ is the component-based
1. Iming: In discrete event simulators, events that
affect state of the system are chronologically discrete event simulator [14,19,21,22]. OMNeT++
ordered into event queue, and event scheduler provides both a noncommercial license, used at
academic institutions or non-profit research
executes them one by one. Continuous
simulators are concerned with modeling a set organizations, and a commercial license, used at "for-
of equations that represent system over profit" environments. OMNeT++ uses building blocks
called modules to implement their simulator. Modules
time [5].
2. Software license: A simulator can be are connected in a hierarchical nested fashion, where
proprietary, or have one of the free each module can contain several other modules.
In the sensor network model of OMNeT++, called
software/open source licenses.
3. Programming Language: Language that SensorSim, modules are used to define a number of
different objects in Figure 2[14].
simulator used.
4. Graphical User Interface (GUI) support: Is it
available and how useful it is?
5. General or specific simulator: They are
examined separately whether it is specifically
designed for wireless sensor network.
IV. SIMULATORS
This paper will present five different simulators.
These simulators were selected based on a number of
criteria including timing, license, key features, merits
Fig. 2: OMNeT++'s Module Implementation
Merits: OMNeT++ provides a powerful GUI. This consumption in WSN. Secondly, every node has to run
strong GUI makes the tracing and debugging much on NesC code, a programming language that is event-
easier than using other simulators. This simulator can driven, component-based and implemented on TinyOS,
support MAC protocols as well as some localized thus TOSSIM can only emulate the type of
protocols in WSN. People can use OMNeT++ to homogeneous applications. TOSSIM is specifically
simulate channel controls in WSNs. In addition, designed for WSN simulation, motes-like nodes are the
OMNeT++ can simulate power consumption problems only thing that TOSSIM can simulate.
in WSN.
Limitation: The number of available protocols is D. J-Sim
not larger enough. In addition, the compatible problem Overview: J-Sim [16,18,19] is a discrete event
will rise since individual researching groups developed network simulator built in Java. This simulator
the models separately, this makes the combination of provides GUI library, which facilities users to model or
models difficult and programs may have high compile the Mathematical Modeling Language, a “text-
probability report bugs. based language” written to J-Sim models. J-Sim
provides open source models and online documents.
C. TOSSIM
Unlike ns-2, however, J-Sim uses the concept of
Overview: TOSSIM [8,15,19,20,23,24], are components, replacing the notion that each node should
actually emulators. Emulators are different from be represented as an object. J-Sim uses three top level
simulators in that they run actual application code. components as shown in Figure 4[16]: the target node
TOSSIM is designed specifically for TinyOS (which produces stimuli), the sensor node (that reacts to
applications to be run on MICA Motes, which is an the stimuli), and the sink node (the ultimate destination
open source operating system targeting embedded for stimuli reporting).
operating system. In 2003, TOSSIM was first
developed by UC Berkeley’s TinyOS project team.
TOSSIM is a bit- level discrete event network emulator
built in Python, a high-level programming language
emphasizing code readability, and C++. People can
run TOSSIM on Linux Operating Systems or on
Cygwin on Windows. TOSSIM also provides open
sources and online documents.
The TOSSIM architecture is made up of a number
of different components as shown in Figure 3[15]:
Angeles Compilers Group. Avrora is implemented in accuracy; Avrora provides more accuracy than
Java, unlike the other two emulators, which are written TOSSIM does with equivalent scales of sensor nodes.
in C. Avrora implements each node as its own thread. Limitation: Avrora fails to model clock drift. It
However, it still runs actual Mica code and is an is 50% slower than TOSSIM.It cannot model mobility.
emulator. Like ATEMU, Avrora runs code in an
instruction-by- instruction fashion. However, the V. ANALYSIS
simulator attempts to achieve better scalability and
speed by not synchronizing all nodes after every Each of these simulators/emulators are
instruction. distinguished using language,timing and other criteria
shown in Table I below.
Merits: Avrora is an instruction-level simulator,
which removes the gap between TOSSIM and A comparative graph of the most prominent
ATEMU. The codes in Avrora run instruction by simulation frameworks according to the criteria of
scalability and abstraction level as shown in fig. 5. In
instruction, which provides faster speed and better
scalability. Avrora can support thousands of nodes this figure, it does not express the maximal feasible
simulation, and can save much more execution time network sizes, but reflects the typical application
domain.
with similar accuracy. Avrora provides larger
scalability than ATEMU does with equivalent
TABLE 1: COMPARISON OF FIVE SIMULATION TOOLS
Simulation tools Simulator or Discret e-Event Simulaions or G UI Open source s General simulator Programming
Emul ator Trace- Driven Simulation and Online or Specific languag e/
documents simulator Platfor m
NS-2 Simulator Discrete-Event Simulat ion No Yes general simula tor C++
OMNeT++ Simul ator Discret e-Event Ye s Yes gener alsimula tor C++
Simulat ion
TOSSIM Emulator Discrete-Event Simulat ion Yes Yes specifically design nesC
ed for WSNs
J-Sim Simulator Discrete-Event Simulation Yes Yes general simula tor Java
Avrora Simulator Discrete-Event Simulation Yes No specifically design Java
ed for WSNs
REFERENCES
[1] M. Ilyas and I. Mahgoub, Handbook of sensor networks:
compact wireless and wired sensing systems, BocaRaton, FL.,
CRC Press, 2004.
[2] S.Yi, H.Min, Y.Cho, J. Hong, “SensorMaker: A Wireless
Sensor Network Simulator for Scalable and Fine–Grained
instrumentation”, computational science and its application-
ICCSA, 2008, Volume 5072/2008, pp. 800-810
[3] D. Curren, “A survey of simulation in sensor networks”,
University of Binghamton project report for subject CS580.
[4] M. Mekni, B. Moulin, "A survey on sensor webs simulation
tools", Proceedings of the 2008 Second International Conference
on Sensor Technologies and Applications, 2008, pp. 574-579.
[5] R. McHaney, Computer simulation: a practical perspective. San
Diego, CA, USA: Academic Press Professional, 1991
Fig. 5: Comparison of Simulators based on Scalability [6] NS-2 URL: http://en.wikipedia.org/wiki/Ns-2,
and Level of Abstraction [7] NS-2URL: http://www.isi.edu/nsnam/ns/
[8] Muhammad Imran, Abas Md Said, Halabi Hasbullah, “A
VI. CONCLUSION Survey of Simulators, Emulators and testbeds for October 4-7,
2004. WirelessSensorNetworks”, Information Techonology
The goals of this paper were to provide (ITSim), 2010 International Symposium in, June 2010,
background scope on a number of different sensor [9] Raj Jain, “Art of Computer Systems Performance Analysis
Techniques For Experimental Design Measurements A Survey
network simulators and present the merit and of Wireless Sensor Network Simulation Tools
limitation of each. Simulators are compared based on [10] “Discrete_event_simulation”,URL:http://en.wikipedia.org/wiki/
different criteria, and comparative results are presented Discrete_event_simulation,
[11] J. Polley, D. Blazakis, J. McGee, D. Rusk, J.S. Baras,
in tabular form. In addition, short descriptions of
ATEMU: A Fine-grained Sensor Network Simulator”,First
simulators are also provided. Since no single simulator Annual IEEE Communications Society Conference on Sensor
under survey is universally applicable to all situations, and Ad Hoc Communications and Networks, Santa Clara, CA,
103 104 10
[12] Yunjiao Xue, Ho Sung Lee, Ming Yang, Kumarawadu, P., [17] Ben L. Titzer, Daniel K. Lee, Jens Palsberg. Avrora: Scalable
Ghenniwa, H.H., Weiming Shen, “Performance Evaluation of Sensor Network Simulation with Precise Timing. Proceedings
NS-2Simulator for Wireless Sensor Networks”, Electrical and of IPSN'05, Fourth International Conference on Information
Computer Engineering, CCECE Canadian Conference on, 22-26 Processing in Sensor Networks, 2005.
April 2007, pp.1372–1375, ISBN: 1-4244-1020-7. [18] Sourendra Sinha, Zenon Chaczko, Ryszard Klempous,
“SNIPER: A Wireless Sensor Network Simulator”, Computer
[13] Valeri Naoumov and Thomas Gross. Simulation of Large Ad
Aided Systems Theory- EUROCAST, 2009, Volume
Hoc Networks. ACM MSWiM, 2003. 5717/2009, pp. 913-920, URL:
[14] C. Mallanda, A. Suri, V. Kunchakarra, S.S.Iyengar, R. Kannan, [19] E. Egea-Lopez, J. Vales-Alonso, A. S. Martinez- Sala, P.
and A. Durresi. Simulating Wireless Sensor Networks with Pavon-Marino, J. Garcia-Haro;Simulation Tools for Wireless
OMNeT++. Sensor Networks”, Summer Simulation Multiconference,
[15] P.Levis, N.Lee, M.Welsh and D.Culler. TOSSIM: Accurate SPECTS, 2005, pp.2-9, URL:
and Scalable Simulation of Entire TinyOS Applications. [20] C. Stevens, C. Lyons, R. Hendrych, R. S. Carbajo, M. Huggard,
Proceedings of SenSys’03, First ACM Conference on C. McGoldrick, “Simulating Mobility in WSNs: Bridging the
Embedded Networked Sensor Systems, 2003. gap between ns-2 and TOSSIM 2.x”, 13th IEEE/ACM
[16] Ahmed Sobeih and Jennifer C. Hou. A Simulation Framework International Symposium on Distributed Simulation and Real
for Sensor Networks in JSim. Technical Report UIUCDCS-R- Time Applications, 2009, ISBN: 978-0-7695-3868-6.
[21] Omnet++”http://en.wikipedia.org/wiki/Omnet.
2003-2386, November 2003.
the local data from nodes to a sink node SamerHanoun et al. [1]models the mobile collector
(gateway or cluster head). and the sensor network as a system-of-systems, and
WSN’s promise several advantages over traditional types of interactions. They presented the framework
sensing method in many ways: better coverage, fault thatassists deploying incongruous mobiles without
tolerance and robustness. The ad hoc nature and deploy priorknowledge about the sensor network.They
and leave vision make them even more attractive in real proposed a model for governing theinteractions
world applications. Several real- world applications are between the sensor network and class ofmobile
being designed and developed that are taking advantage elements acting as data collectors. They showed
of this new technology. amethodology of structuring the sensor network
For example:-In tree for fire detection [13], In forconveying collection request to the mobiles
building for fire crew assistance [14], In wine grape withoutknowing their positions or number. Two motion
vineyards for improving the quality of crop and the heuristicsare presented to guide the collection tour of
performance of the land and in many other applications the mobileelement. They argue that theproposed
such as military application, monitoring toxic zones, framework is both practical and simple for
agriculture, healthcare, industrial automation and providingloose coupling between the network and the
volcano monitoring [15]. mobilecollectors.
A.S.Poornima et al. [2] proposed a simple
authentication scheme for heterogeneoussensor network
which uses mobile agents forefficient data collection.
There proposed scheme is used torecognize and find
malicious nodes acting as mobile agents. Also attaining
confidentiality of the data collected using simplekey
derivation technique which allows a cluster head
toencrypt the data every time using a different key
which canbe easily derived by the base station. Mobile
agents is used to collect the data is becomingpopular
Fig. 1: Wireless Sensor Network in wireless sensor networks. Regarding security there
were challenges when mobile agents are used for data
A. Characteristics
collection.They have proposed a simple
The main characteristics of a WSN include: authenticationscheme which is derived using the tree
• Power consumption constrains for nodes using based key managementscheme. The proposed scheme
batteries or energy harvesting identifies malicious MA and replay messages. Through
• Ability to cope with node failures using refreshed secret key Pi transferred data is
• Mobility of nodes encrypted. The secret key is known only to CH and
• Communication failures base station. Hence the scheme is resilient
• Heterogeneity of nodes tocompromise of MA.
• Scalability to large scale of deployment Hongjoong Sin et al. [3] proposes an energy
• Ability to withstand harsh environmental efficient agent-based structure in wireless sensor
conditions networks. They adopt biologicallyinspired approaches
• Ease of use for wireless sensor networks. Agent
operatesautomatically with their behavior policies as a
• Power consumption
gene. Agent agglomeratesother agents to reduce
communication and gives high priority to nodesthat
III. ENERGY EFFICIENT APPROACHES IN WIRELESS
have adequate energy to communicate. Agent behavior
SENSOR NETWORK
policies areoptimized by genetic operation at the base
Wireless sensor nodes have limited energy station. Each agent selects a next-hop node with
resources. Advance in battery technology has been very neighbor information andbehavior policies. There
much limited compared to the development of sensor proposed framework provides self- configuration, self-
networks. Energy conservation is essential in many healing, self-optimization properties to wireless sensor
scenarios due to the impossibility of extending or nodes. The sensor node generates an agent, when the
replacing the battery power. In addition, the multi- hop node has the sensing data. Theagent contains the
routing in a typical sensor network from the layered or sensing data and channels the data toward thebase
clustered architecture has introduced a large portion or station hop by hop. The base station selects elite agents
relay traffic which depletes the power quickly as in a thatthey have more appropriate behavior policies in
traditional wireless network. Deployment of mobile currentnetwork situation by periods. The base station
sensors in a wireless sensor network is an energy propagateselected elite agents to each node. Agents in
consuming process and it should be carefully designed. each node inheritbehavior policies of elite agents.
Edwarrd J.Coyle et al. [4] prooposed an ennergy aconntrollable infrastructure in the sensor field,
f which
efficient hiierarchical cluustering schem me to enable nodes
n acts aslow-priorrity data coollection ageent. There
transmissioon within loong distance. The sensorss are propposed approacch can providee ample energ gy saving in
organized into groups and communnicate inform mation this class of sensoor application compared totthe classical
only to cluster
c heads and then the t cluster heads
h multtihop approachh.
communicaate the accuumulated infformation too the SumitChaudhhary et al. [8] proposed th he efficient
processing center. The processing
p cennter determinees the mech hanismof energy efficiennt techniquess for data
final estim
mates of the parameters in question usinng the aggrregation in WSN using prinnciples like gllobalweight
informationn transmitted by the clusteer heads. Sincce the calcuulation of noodes, data colllection for cluster
c head
sensors area now traansmitting daata over sm maller and data aggreegation technniques using g datacube
distances ini the accumuulated environment, the ennergy aggrregation.
spent in thhe network wiill be much lesser as compaare to
the energyy spent whenn every senssor communiicates A. Data Cube Agggregation
directly to the informatioon processingg center. It is a multidimensionnal approach h for data
HaniehhAlipour et al. [5] propposes an effi ficient aggrregation. The values
v are stoored in separaate cell of a
approach for
f data collecction in wireleesssensor netw works data cube, each phase
p of cubee is divided innto separate
by introducing Memberr Forward Lisst. This listinccludes rowss & columnss and each value & nod de such as
the nodes with
w highest priority
p for foorwarding the data. conssumption, baandwidth, M MRIC, RSSII etc are
When a noden t next nodewith
dies, thee list select the repreesented at the beginning of rows.
higher priority. The benefit
b of thiis node is thhat it
secures thee algorithm from repeating when a node dies.
Member Foorward List decreases poweerconsumptionn and
latency in wireless senssor networks.T The list helpps the
other nodees to find thee path for dataa forwarding when
a previous forwarder noode has failedd. They introdduced
an efficiennt protocol to encounter foorwarder nodde for
saving ennergy. In thiis protocol they t usedMeember
Forward List
L (MFL) to find an adequuate and theshoortest
path for forrwarding dataa to the sink.
L. Beenini et al. [66]analysed thhe efficient power p
saving in the
t micro-conntrolled unit. They T discusseed the
choice of altered
a processsors for partiicular applicattions.
The dynam mic power mannagement (DP PM) approachh can
be used to shut down some s elementts in a sensor node
to save ennergy when itt becomes ouut of action for a
while. Thee software paart in a sensor network such as
the operrating system m, networkk communiccation
protocols, and applicattion protocolls should alsso be
optimized to abate the use of energgy by abatingg the Fig. 2: Data Cube Agggregation[8]
unrequiredd broadcast messages
m for thhe handshakinng or
other redunndant design in i a traditionaal wireless nettwork There paperr widely accclaims the improved
protocols. techn nology for energy effficient techn niques for
Amar Rasheed et al. [7] proopose an ennergy- dataaaggregation and
a collectionn in WSN. There
T work
efficient hybrid
h data collectionarch
c hitecture baseed on provvides the accuurate usage oof battery andd lowpower
controllablly movable infrastructurre for aclasss of conssumption so that the usser can sen nd multiple
applicationns in which sensor networks n proovide messsages in limitted resources.. Theparameters that are
bothlow-prriority and high-priority data. d High-priority usedd manage the cluster head ggeneration, an nd the node
dataneed a data deliveryy arrangemennt with low laatency selecction methods sothat the message can n be easily
and highfiidelity. Concuurrently low-priority data may transsferred undeer such circcumstances with right
tolerate highlatencydat
h ta delivery. There apprroach decission usingprinnciples like gglobal weight calculation
accomplishhed the design of anetw work that braace a of nodes, data collection for cluster head and
hybrid daata delivery arrangemennt toenhance the dataaaggregation techniques using data
d cube
network performance
p a
and abate total networkennergy aggrregation.
usage. In there system m design two delivery schhemes Mohammad HosseinAnissi et al. [9]proposed
[
aredeployeed for purposees of comparisson. The first is the energ gy-efficient mechanism, the most appropriates
a
classical add hoc approaach to deliverr high-priorityy data hopss for data foorwarding wiill be selecteed and the
withhigh fidelity
fi and loow latency. Thhe second preesents lifetiime of the whhole network will be increeased. They
proppose an energy-efficie
e ent in-Netw work data
aggregation approach in WSN. The proposed approach data from sensors, the sink movement iscombined with
uses the benefits of both cluster based and tree based three data collection strategies: a passive, amulti-hop
accessions. The whole network consist clusters of the and a limited multi- hop.
same size. Each node is related to a routing sub tree and Raghavendra et al. [12] have proposed power-
each sub tree floods a cluster and the root node of each efficient gathering for sensor information systems
sub tree is the head node of the associated cluster. The (PEGASIS) [16] to transmit the collected data from
energy usages in wireless communication is equal to sensor nodes to the sink node. It is based on the
the square of distance between two nodes in assumption of a relatively static network topology in
transmission. All the nodes transmit their data to their which each node knows the location of other nodes in
neighbor instead of their cluster head. Therefore, the the network. The node which was selected as a leader
transmission distance is abated and the energy usages of was within the transmission range to the sink node in
each node, each group and the average energy usages of one hope. A chain is constructed from the utmost nodes
the whole networks is reduced and the network lifetime to the sink before the data is transmitted. When leader
is elevated.
transmit the message to the sink then the message will
Tao Shu et al. [10] develop mechanisms that
pass from one node to the next node.
generate randomized multipath routes. Under there
designs, thepaths taken by the “shares” of different
IV. CONCLUSION
packets change over time. So even if the routing
algorithm becomes known to the opponent,the Efficient energy consumption and conservation is a
opponent still cannot pointout the routes cross over by key problem in the sensor networks, lots of work has
each packet. Besides irregularities, the generated been done for energy efficiency but less attention has
paths are also highly distributive and energy efficient, been paid in alternative approaches for energy
making them quite capable of mislead black holes. efficiency. This study is trying to conduct a survey of
Theytentativelycheck out the security andenergy various approaches which could be applied in wireless
performance of the proposed schemes. They also sensor networks considering the efficiency and
formulate an optimization problem to decrease the end- cooperating of mobile sensors to collect information
to-end energyconsumption under given security and make decisions.
compulsions. Wide simulations are conducted to verify
the validity of our mechanisms.They propose REFERENCES
aarbitrary multipath routingalgorithm that can
[1] SamerHanoun and Saeid Nahavandi., “A Practical Framework
overcome the above problems. In theirapproach, for Data Collection in Wireless Sensor Networks,” Proceeding
multiple paths are computed in aarbitrary way each of IEEE 2008-978-1-4244-2173-2/08.
time an information packet needs to be sent, suchthat [2] A.S. Poornima and B.B. Amberker., “Agent Based Secure Data
the set of routes taken by various shares of Collection in Hetrogeneous Sensor Networks,” IEEE
Proceeding 2010.Second International Conference on Machine
differentpackets keep changing over time. As a result, a Learning and Computing.
large numberof routes can be conceivably generated for [3] Hongjoong Sin, Jangsoo Lee, Sungju Lee, SeunghwanYoo,
each source anddestination. To interrupt different Sanghyuck Lee, Jaesik Lee, Yongjun Lee, and Sungchun Kim,
packets, the opponent hasto accord or jam all possible “Agent-based Framework for Energy Efficiency in Wireless
Sensor Networks,” World Academy of Science, Engineering
paths from the source tothe destination, which is and Technology 46 2008.
practically not possible. [4] SeemaBandyopadhyay and Edward J.Coyle., “An Energy
IoannisChatzigiannakis et al. [11] proposed the Efficient Hierarchical Clustering Algorithm for wireless Sensor
idea of having a sinkmoving in the network area and Networks,” IEEE INFOCOM 2003.
[5] HaniehAlipour and AlirezaNemaney Pour, “An Efficient
collecting data from sensors.They propose four DataCollection Approach forWireless Sensor Networks,”
characteristic movability patterns for the sinkalong with World Academy of Science, Engineering and Technology
different data collection strategies. There data 56 2011.
demonstratethat by taking advantage of the sink’s [6] T. Simunic., L. Benini., and G. D. Micheli., “Energy- Efficient
battery Management,” Proceedings of IEEE INFOCOM 2000,
mobility, wecan significantly abate the energy spent in vol 2, pp. 396-403, March 2000.
hand on trafficand thus greatly extend the lifetime of [7] Amar Rasheed., Rabi Mahapatra., “An Energy-Efficient Hybrid
the network. They propose and inquiresink mobility as Data Collection Scheme in Wireless Sensor Networks,”
a method for efficient and robust datadelivery in Proceeding IEEE 2007.ISSNIP 2007.
[8] Sumit Chaudhary., Neha Singh., AvinavPathak and A.K Vatsa.,
wireless sensor networks. They propose four “Energy Efficient Techniques for Data aggregation and
mobilitypatterns for the sink, mostly randomized (such collection in WSN,” International Journal of Computer Science,
as the simplerandom walk, biased random walks and Engineering and Applications (IJCSEA) Vol.2, No.4,
walks on spanningsubgraphs) as well as predictable August 2012.
[9] Mohammad HosseinAnisi,. Abdul Hanan Abdullah.,
mobility (moving ona straight line or cycle). These ShukorAbdRazak.,“Energy-Efficient Data Collection in
patterns assume and exploitdifferent degrees of Wireless Sensor Networks,”Wireless Sensor Network, 2011,
freedom, simplicity and networkknowledge. To get 3, 329-333. doi:10.4236/wsn.2011.310036 Published Online
October 2011.
[10] Tao Shu., Marwan Krunz., and Sisi Liu., “Secure Data [13] G. Boone, “Reality mining: Browsing reality with sensor
Collection in Wireless Sensor Networks Using Randomized networks”, Sensors online, vol.21, no. 9, Sept 2004.
Dispersive Routes,” IEEE TRANSACTIONS ON MOBILE [14] Fire information and rescue equipment [online](June,2008)
COMPUTING, VOL. 9, NO. 7, JULY 2010. [15] Deb, B., Bhatnagar, S., Nath and B., “Reinform: Reliable
[11] Ioannis Chatzigiannakis., Athanasios Kinalis., Sotiris Information Forwarding using multiple paths in sensor
Nikoletseas., “Sink Mobility Protocols for Data Collection in networks,” in IEEE International Conference on Local
Wireless Sensor Networks,” MobiWac’06, October 2, 2006, Computer Networks (LCN’03), 2003.
Torremolinos, Malaga, Spain.Copyright 2006 ACM
159593488X/06/0010.
[12] S. Lindsey and C. S. Ragjavemdra., “PEGASIS: Power-Efficient
gathering in Sensor Information Systems,” Proceeding of IEEE
ICC 2001, Vol. 3, pp. 1125-1130, June 2001.
Abstract—Data mining is the task of discovering experts should work together on the same data mining
interesting patterns from large amounts of data. There are task.These observations provide motivation for the
many data mining tasks, such as classification, clustering, developmentof a methodology for collaborative data
association rule mining and sequential pattern mining. mining. Our
Sequential pattern mining finds sets of data items that
occur together frequently in some sequences.
pointof departure is that groups with different
Collaborative data mining refers to a data mining expertise who aregeographically distributed should be
settingwhere different groups are geographically able to collaborate on acertain problem, thus jointly
dispersed butwork together on the same problem in a achieving better results thanany of them could
collaborative way. Such a setting requires adequate individually.In the context of theEuropean SolEuNet
software support.Group work is widespread in education. project, ideas have evolved about whatfunctionality
The goal is to enable the groups and their facilitators to such an environment should offer, resulting ina
see relevant aspects of the group’s operation and provide proposal for a collaborative data mining methodology
feedbacks if these are more likely to be associated with andsupporting system called RAMSYS [3] and
positive or negative outcomes and where the problems are.
We explore how useful mirror information can be
animplementation using the groupware system Zeno
extracted via a theory-driven approach and a range of [2]. The paper is structured as follows. In Section 2 we
clustering and sequential pattern mining. In this paper we discussRAMSYS and Zeno. In Section 3 we describe
describe an experiment with a simple implementation of ourcollaborative data mining experiment and the
such a collaborative data mining environment. The problemsencountered, and in Section 4 we propose and
experiment brings to light several problems, one of which comparepossible solution. Section 5 concludes.
is related to model assessment. We discuss several possible
solutions. This discussion can contribute to a better
understanding of how collaborative data mining is best II. DATA MINING IN COLLABORATIVE MANNER WITH
organized.
PROPOSED TECHNOLOGIES
Keywords: Data Mining, Sequential Pattern Mining,
Collaborative data mining, Data Preparation. We set our primary goal for the data mining as
providingmirroring tools that would be useful for
I. INRODUCTION helping improvethe learning about group work. This
goal is realistic in thecontext of the highly complex and
Group work is commonplace in many aspects of variable nature of longterm, smallgroup activity,
life, particularly in the workplace where there are many especially wherethe learnersundertakea diverse rangeof
situations which require small groups of people to tasks, suchas creating asoftwaresystem for an authentic
worktogether to achieve a goal. For example, a task that client.Data mining is about solving problems by
requiresa complexcombination of skills may only be analyzing dataalready present in databases [4]. Problem
possibleif a group of people,each offeringdifferent solving can be codified and a procedure or
skills,canwork together.To take justone other example,it methodologycan be developed. There are one
maybenecessarytodrawonthe combinedefforts ofa methodology in data mining that is used for industrial
grouptoachieve a task in the time available [1]. Many process we called Cross Industrial Standard Process for
different approaches to data mining exist. They have Data Mining, CRISP-DM [5]. This technique is used to
arisen from different communities (databases, statistics, reduce the data mining problems into six-related phases
machine learning. Thus, data mining nowadays is of (i) Data Understanding (ii) Data Preparation
performed by people with highly different backgrounds, (iii) Business Understanding (iv) Modeling (v)
each of whom have their preferred techniques. Very Assessment and (vi) Deployment. These phases,
few people are experts in all these domains, so to get although presented in a linearmanner, have many cycles
the most out of a data mining process; ideally multiple and feedback loops connecting the phases. Often, effort
expended in one phase highlights the need for further visitors and trying to link properties of thesevisitors to
work in a prior, previously consideredcomplete, phase. the procedures they followed, predictive modelscould
The Cross Industrial Standard Process for data be built that estimate the demand for certain
mining method is extended by RAMSYS technology in facilitiesbased on known properties of future visitors.
the methodology for distributed teams who collaborate Thus the data mining task can concisely be
in adata mining project.The aim is to combine the great described as follows: given a set of visitor descriptions
range of expertise available in the data miners to affect that will arriveduring a certain week, estimate how
morevaluable solutions to the data mining problem. The many of these visitorswill need to follow each of some
RAMSYS methodology attempts to achieve the available procedures.
combination of a problem solving methodology,
knowledgesharing, and ease of communication. It is A. Data Mining Method in Collaborative Approach
guided by the following principles [3]: it should enable Group work is used in this scenario, so we make 4
lightmanagement, it should allow collaborators to start groups in college students. Those are described as: CSE
and stop any time and leave them problem solving (Computer Science Engineering), ECE (Electronics
freedom, it should provide efficient knowledge sharing Communication Engineering), EE (Electrical
and security. The RAMSYS efforts have focused on Engineering), ME (Mechanical Engineering). CSE
supporting the Data Preparation and Modeling phase in served as contact withthe end user (the health farm).
a remotecollaborative setting; here we focus on the Following the RAMSYS methodology implies
Assessment phase. There is the main requirement of the following the CRISP-DM methodology, hence we here
RAMSYS scheme is the emphasizing and availability briefly describe the efforts according to the different
of the current preeminent understanding [6] of the data phases Phase 1 (business understanding) involved
mining problem. This has been implemented using the becoming familiar with the data mining problem, which
academic groupware platform Zeno [2], by providing was done byall groups separately. During Phase 2 (data
coordination, collaboration, communication and understanding) several groups explored the data using
awareness. The provisions of these features are visualizationtechniques, association rule discovery, etc.
achieved by utilizing (new)features in Zeno including and publishedtheir results on Zeno. In Phase 3 (data
task management, resourcemanagement, and discussion preparation) themain effort consisted of data
sections. The RAMSYS methodology has been trialed transformations. As theoriginal database consisted of
(in part or infull) on several data mining projects, one of multiple tables, this involvedto some extent
which is the SPA project described in the subsequently computation of aggregate functions. Data
section. transformations were performed mainly using CSE’s
Sumatra TT tool [7].
III. A PRACTICAL APPROACH WITH COLLABORATIVE Mainly we focus on Phases 4 and 5: modeling and
DATA MINING assessment. Concerning modeling, a wide variety of
There are numerous health farms, provide a number approaches was taken by the different groups: support
of facilities that can be used by its visitors. More vector machines (ECE), neural nets (ECE, EE),
specifically, upon their arrival visitors are prescribed linearregression (EE, ME), instance based learning (EE,
certainprocedures to follow during their stay at the spa, CSE), decision trees (EE, CSE, ME), etc. Besides the
as well as aschedule for them. The number of people different algorithms, approaches also differedin the
that cansimultaneously make use of certain facilities is version of the data set that was used (these
limited. Thus the spa is faced with a scheduling task: versionsresulting from different data transformations).
given the procedures that newly arrived visitors need to There is an intense feedback from 5 to 4: based on
follow and the limited capacity of certain facilities, model assessment, data miners wish to change their
create a suitableschedule. The “SPA problem” was model building approach and go through Phases 4 and 5
offered to the Sol Eu Netconglomerate by a health farm. once more. In the collaborative setting, the feedback
In practice there is insufficient information to solve should not remain with in one group but flow to all
this scheduling task for the following reason. Visitors groups for which it is relevant.
stay for several weeks and a schedule for their whole B. Assessment of the Collaborative Data
period of stayis made, but during their stay new visitors Mining Process
will arrive. Whilesome information about these new
visitors is available inadvance (such as time of arrival, In this Collaborative data mining, assessment is
age, sex …) the procedures they need to follow will be ambiguous. The end-user found the results interesting
known only at the time of their arrival. The best one can and useful [8].An awful thing is that the added value of
do is to estimate the demand for the facilities for the collaboration of different groups on this task was much
near future, and use theseestimates for producing smaller than hoped. The most notable collaboration was
schedules for the current patients. It is here that data that the results of data transformations performed by
mining comes in: by mining a data base of previous one group were used for modeling by another group.
This is inline with the kind of collaboration that one should have a kind of model assessment server to
RAMSYS promotes, but it is only a minimal version of whichgroups send the models they have produced, or
it. To achieve more intensive collaboration, several thepredictions produced by their models. When a
processesmust be made more efficient. The CRISP-DM groupdecides they are interested in some specific
process is iterative, consisting of many steps and cycles. criterion, theyshould be able to add the criterion to the
Ifcollaboration is to happen at the level of a single step, central assessmentserver and immediately see the scores
itneeds to happen very efficiently. To make this of all earliersubmitted models on these criteria. In the
possible, information exchange should be made more next section weexplore this direction further.
efficient and synchronization should be improved. The
information flow between groups was often hampered IV. CENTRIC ASSESSMENT APPROACH
because documentation of results was too concise, too
In our idea, data mining groups (“clients”) should
extensive, oreven both (groups being flooded with
information from colleagues without being able to find sendforecasts or even the models themselves to a
the most relevantinformation in there). As groups do “model assessment server”, which is responsible for the
assessmentof the prognostic model and automatically
not always have theright resources available at the right
time, it may take awhile before a group reacts to results publishes theresults.
from other groups. The solutions to these problems are
to be found both at the technical and management level
(e.g. defining strict formats for exchanged documents
so that relevant information iseasier to identify).
Second, process that needs to be made more
efficient is comparative assessment of models. In order
to comparedifferent models, they must be evaluated
according to thesame criteria. The original RAMSYS
methodology proposed to determine an assessment
criterion in advance so that each group can assess their
models according to thiscriterion. The SPA experiment
revealed several problems with this proposal. Firstly, it
may be difficult to propose agood assessment criterion
in advance, and the preferredassessment criteria may
change over time, because insight in what are good and
bad criteria typically develops during the knowledge
discovery process. E.g., in the SPA experiment, visual
data analysis revealed strong outliers. These turnedout
(after discussion with the end user) to be related
tunavailability of certain procedures due to maintenance
andwere therefore irrelevant, but they strongly
influencedcertain error criteria and needed to be left
out.
In other approach, one criterion may not be
sufficient. Differentcriteria measure different properties,
all of which may berelevant, see e.g. [9]. It is more
realistic to talk of a set ofcriteria, instead of a single
one. And finally, subtledifferences in the computation
of certain criteria, the data setfrom which they are
computed, the partitioning used for cross-validation can
make the comparison unreliable.
Due to the rate at which criteria may change, the
number of criteria, and the care that must be taken when
implementing them, it is unrealistic to expect that the
different groups will continuously use the right criteria.
An assessment scheme is needed in which criteria can
flexibly be changed or added and it is guaranteed that
every group uses exactly the sameversion of a criterion,
without too much overhead. Fig. 1: Overview of Centric Model Assessment for Collaborative
We suggest centralized model assessment. Instead Data Mining
ofhaving all different groups assess their own models,
There are several levels of communication are soon as a single implementation of the newcriterion is
possible. An inductivesystem typically has a number of available. Comparability reflects the trust in the
parameters; for a given setof parameters values the comparability of the results, which is higher when a
system implements a function I: 2X×C→ (X → C) that singleimplementation is used. Student transparency
maps a dataset (a subset of the universe of labeled refers to the overhead for the data mining groups when
instances X×C with X the instanceuniverse and C the set some option isadopted. In Option 1 it is highest, in
of target values) onto a function M (apredictive model) Option 4 lowest because the user need only submit I
that in turn maps single instances ontosome target (induction system+parameters) and all testing is then
value. One has the option to submit theinductive done automatically. In Options 2 and 3 the user needs to
function I; the model M learnt from a given dataset T; implement e.g. cross validation according to given
or a set of predictions for some data set S, P = folds. Finally flexibility of assessment is lowest for
{(e,M(e))|e∈S}. In all cases the server should be able Option 2 because here the criterioncannotinvolve the
toderive from the submission a score on one or modelitself (complexity, interpretability but only it
moreevaluation criteria, which we assume to be a spredictions.
function (M,P). The original RAMSYS procedure Option 1 is the current mode of operation with in
corresponds to afourth option, communicating s = Sol Eu Net. Option 2 provides significant advantages
c(M,P). over Option 1 andis still easy to implement. Option 3
A schematic overview of these options (in reverse imposes the challenge that a good model description
ordercompared to above) is given in Figure 1. It is language and an interpreter for it need to be available.
assumed thatI consist of a combination of a machine A reasonable choice for such alanguage would be
learning tool andparameter settings, so Iis the result of PMML [10], which is already being proposed as a
tuning the tool with the parameters. Using I a model M common language for representing models; it handles a
is built from a trainingset, this M is used to predicted reasonable variety of types of models and thereexist
labels for a test set S, fromthese predictions a score s is visualizes for them. If PMML is going to be used
computed using the evaluationcriterion c. In the case of
anyway in a collaborative data mining system, an
a cross-validation the process ismore complicated but
interpreter for PMML models would be sufficient to
the same basic scheme is valid: different models Mare
cater for a widerange of different model assessment
then built from different trainingsets to produce one set
of predictions P. criteria.
In data mining, we describe a data set in a table, in Option 4 is the most powerful one but seems least
which various states are identified as following shown feasible. There are different suboptions: (4a) all model
table. building systems are translated into a single common
language; (4b) the central model assessment server has
TABLE 1: CHARACTERISTICS OF DIFFERENT OPTIONS the necessary interpreters for the different languages in
States 1 2 3 4 which inductive systems, data preprocessing systems,
Question difficulty L L M H etc. are programmed;(4c) the server has its own
Efforts cost L H M M versions of the inductive systems, and all that is actually
Result availability L H H H submitted is an identifier of the system to be used and a
Comparability M H H H
Student transparency H M M L
list of parameters. Option 4c isquite feasible but has the
Flexibility of assessment H M H H disadvantage that only the systems and versions
available at the server can be used.
Table 1 precise some characteristics of the four In the short term, we believe the most realistic
options. In the table H, M and L refer to High, improvement to RAMSYS corresponds to Option 2,
Mediumand Low respectively. Question difficulty refers which is easy toimplement and presents a significant
to the questionsthat are needed for examination. improvement over the current mode of operation. In the
Options 3 and 4 impose the challenge of developing longer run, assuming that PMML is general enough to
relatively complexlanguages and interpreters for them describe any kind of model that could be submitted and
(e.g., when submitting amodel M the server needs to be that interpreters are available, itseems desirable to shift
able to compute thepredictions M makes on some test to Option 3.
set). Summarizing, centralized model evaluation
Efforts cost is low when result just a score, high reducesworkload; increases confidence in comparisons
whencommunicating a (possibly large) set of between systems; guaranteesavailability of all criteria
predictions, andmedium when communicating for all models; reduces the time needed to obtain scores
functions. Result availability refers to how fast the on newcriteria; and a dds flexibility w.r.t. defining new
scores of different models for a newcriterion are made criteria. All of these contribute to the added value that
available to everyone. It is low forOption 1 (groups collaborative data mining can have over the non-
need to implement the new criterionthemselves); for collaborative approach.
other options new scores are automaticallycomputed as
V. CONCLUSION [3] Jorge, A., Moyle, S., Voß, A., and Richter, G. (2002). Remote
collaborative data mining through online knowledge sharing.
We performed mining of data collected from students PRO-VE ’02, 3rdIFIP Working Conference onInfrastructures
working in teams and using a collaboration toolin a for Virtual Enterprises, Portugal.
[4] Witten, I., Frank, E. (1999). Data mining: practical machine
one-semester. Our goalwas to support learning group learning tools and techniques with Java implementations.
skills in the context of astandard state-of-the art tool. Morgan Kaufman, San Francisco.
Collaborative data mining, as promoted by and used [5] Chapman, P., Clinton, J., Kerber, R., Khabaza, T., Reinartz, T.,
with in the Sol Eu Net project, is not a trivial enterprise. Shearer, C. and Wirth, R. (2000). CRISP-DM 1.0: stepby-step
data mining guide.
In order forit to work well, a highly tuned supporting [6] Voß, A., Gärtner, T. & Moyle, S. (2001). Zeno for rapid
environment is needed. An experiment with collaboration in data mining projects. In Proc. of the
collaborative data mining, following the RAMSYS ECML/PKDD-01 Workshop on Integration of Data Mining,
methodology as much as possible, indicated the need Decision Support and Meta-Learning, pp. 43–54.
[7] Aubrecht, P. &Kouba, Z. (2001). Metadata driven data
for more efficient and flexible model evaluation. Our transformation. In Proc. of the 5th World Multi-Conference on
answer to this is centric assessment model, of which we Systemics, Cybernetics and Informatics.
have presented and compared several versions. The [8] Stepankova, O., Lauryn, S., Aubrecht, P., Klema, J., Miksovsky,
conclusion is that significant improvements over the P., Novakova, L., &Palous, J. (2002). Data mining for resource
allocation: a case study. Intelligent Methods for Quality
approach used for SPA can easily be obtained, while
Improvement in Industrial Practice, 1.
implementing an ideal system will need some more [9] Köpf, C., Taylor, C., & Keller, J. (2001). Multi-criteria meta-
work. learning in regression. In Proc. of the ECML/PKDD-01
Workshop on Integration of Data Mining, Decision Support and
REFERENCES Meta-Learning.
[10] Wettschereck, D., & Müller, S. (2001). Exchanging data mining
[1] E. Salas, D.E. Sims, and C.S. Burke, "Is There a "Big Five" in models with the predictive modelling markup language. In Proc.
Teamwork?," Small Group Research, vol. 36, pp. 555-599, of the ECML/PKDD-01 Workshop on Integration of Data
2005. Mining, Decision Support and MetaLearning, pp. 55–66.
[2] Gordon, T., Voß, A., Richter, G., Märker, O. (2001). Zeno:
Groupware for discourses on the internet. KünstlicheIntelligenz
15: 43–45.
a switch that's part of the telephone system's network network security strategy requires identifying threats
of switches. and then choosing the most effective set of tools to
If I place a call to you, I pick up the receiver, and combat them.
dial your number. This number specifies which central
office to which to send my request, and then which A. Comparison with Information Security
phone from that central office to ring. Once you answer The terms network security and information
the phone, we begin talking, and our session has begun. security are often used interchangeably, however
Conceptually, computer networks function exactly the network security is generally taken as providing
same way. protection at the boundaries of an organization, keeping
It isn't important for you to memorize the ISO/OSI the bad guys (e.g. hackers, script kiddies, etc.) out.
Reference Model's layers; but it's useful to know that Network security systems today are mostly effective, so
they exist, and that each layer cannot work without the the focus has shifted to protecting resources from attack
services provided by the layer below it. or simple mistakes by people inside the organization,
e.g. with Digital Leak Protection (DLP). One response
Application to this insider threat in network security is to
compartmentalize large networks, so that an employee
would have to cross an internal boundary and be
Presentation
authenticated when they try to access privileged
information. Information security is explicitly
Session concerned with all aspects of protecting information
resources, including network security and DLP.
a ‘reversee path’ for the monitooring and reemote association of a large quantiity of inform mation, and
adjustmentt of equipmennt in the field. imprroved personnalization. Convergence with other
Exampples of comm munications suubsystems innclude netw
work (enterpriise, fixed) seervices will come
c about
the Defensse Communicaations System (DCS). throuugh the high session
s data rrate. Machine--to-machine
A taactical com mmunications system is a transsmission will involve two basic equipm ment types:
communicaations system m that (a) is usedu within, or in senso ors (which measure
m arametters) and tagss (which are
i designed to meet
direct suppport of, tacticaal forces, (b) is geneerally read/wrrite equipmennt). It is exp pected that
the requireements of chhanging tacticcal situationss and userss will requiree high data raates, similar to
t those on
varying environmentaal conditions, (c) proovides fixedd networks, for
f data and streaming applications.
a
securable communicatioons, such ass voice, data,, and Mob bile terminal usage (lapptops, Person nal digital
video, amoong mobile usersu to facilittate commandd and assisstants, handheelds) is expeccted to grow w rapidly as
control witthin, and in suupport of, tacttical forces, annd (d) they become morre users frienndly. Fluid high h quality
usually reequires extrem mely short installation
i t
times, video o and netw work creactiviity are impo ortant user
usually onn the order of o hours, in ordero to meeet the requ
uirements. Keey infrastructuure design reequirements
requiremennts of frequentt relocation. inclu
ude: fast respoonse, high sesssion rate, hig
gh capacity,
An Emergency
E coommunicationn system is any low user chargees, rapid retuurn on inveestment for
system (typpically compuuter based) thhat is organizeed for operrators, investm
ment that is inn line with thee growth in
the primaary purpose of supportinng the two way demaand, and simpple autonomouus terminals.
communicaation of emerrgency messaages between both As a compariison, the expeected best perfformance of
individualss and groups of individuaals. These sysstems 3G is around 10 1 bit/s/Hz/kkm2 using High H Speed
are comm monly designned to inteegrate the cross- c Dowwnlink Packet Access (H HSDPA), Mu ultiple-Input
communicaation of messsages between are varietty of Multtiple-Output (MIMO),
( etc. No current technology is
communicaation technoloogies. capaable of such peerformance.
A. Key 4G Technnologies
IV. COM
MMUNICATIO
ON
Some of the key technologies required for 4G are
brieffly described below:
b
V. OFDM
MA
Orthogonal Frequency Division MultiplexingM
(OFD DM) not onnly provides clear advaantages for
physsical layer performance, buut also a fram mework for
imprroving layer 2 performaance by pro oposing an
addittional degreee of free-dom m. Using OD DFM, it is
possible to exploiit the time doomain, the spaace domain,
the frequency doomain and evven the code domain to
optimmize radio chhannel usage.. It ensures very v robust
Fig. 2 transsmission in multi-path
m envvironments with
w reduced
receiiver complexiity. As shown in Figure 3, the t signal is
split into orthogonal subcarrierrs, on each off which the
signaal is ‘narrow wband’ (a feew kHz) and d therefore
immmune to multi-ppath effects, pprovided a gu uard interval
is in
nserted betweeen each OFD DM symbol. OFDMO also
provvides a frequuency diversity gain, imp proving the
physsical layer perrformance. It is also comp patible with
otherr enhancemennt technologies, such as smaart antennas
and MIMO. OFD DM modulatioon can also bee employed
as a multiple acceess technologyy (Orthogonall Frequency
Division Multiplee Access; OFD DMA). In thiss case, each
OFD DM symbol caan transmit infformation to/fr from several
Fig. 3 userss using a diffferent set of suubcarriers (su
ubchannels).
This not only provvides additionnal flexibility for
f resource
The evolution from m 3G to 4G will be driveen by
alloccation (increaasing the cappacity), but also enables
services thhat offer betterr quality (e.g.. video and soound)
thanks to greater
g bandw width, more soophistication in i the crosss-layer optimiization of radiio link usage.
268 Bhag
gwant Institute of
o Technology, Muzaffarnagar,
M Uttar Pradesh, India
Co
ommunication S
System and Netw
work Security
VI. CONCLU
USION
Abstract—Today organizations are performing most major Challenge in any cloud computing infrastructure,
of the work with the help of computers and internet. because it is necessary to ensure that only authorized
Online application multiplies the business of the access is permitted and secure behavior is accepted. [2].
organization with automation and accuracy. Cloud 04 (Four) Deployment models are available in the
provides the solution of nonstop working of the systems
and application around the world and around the clock.
cloud environment. Customer can choose any model
Organizations choose cloud as because it is very helpful in depending upon their requirement and budget
fast deployment of the application, secure access and
without much more costing on computing environment.
A. Private Cloud
With the use of hybrid hosting organizations can take 1. The entire cloud based infrastructure is used
benefits of public and private cloud and get and scalable only for one customer. This is a type of Cloud
and elastic infrastructure for their dynamic modern hosting that is usually chosen and managed by
applications. Customer can choose the deployment model
according to their requirement and according to their
an enterprise. In private cloud the entire cluster
budget. of Cloud servers is leased by the enterprise,
which can be managed by them or the host
Keywords: Hybrid Hosting, Conventional Hosting, depending on a case to case basis.
Benefits of Hybrid Hosting / Cloud 2. This is a perfect solution for businesses who
wishes to use the infrastructure for carrying out
I. INTRODUCTION companies internal operations. This type of
Cloud computing is very hot buzz today. The Cloud can be considered to be most secure
number of industries using cloud is increasing day by amongst others as there is no outsider using the
day. Cloud computing basically came from cluster.
telecommunication companies when the virtual private 3. The enterprises can prefer to manage the
network (VPN) technologies and services were cluster on their own, where the hardware
developed to provide security and lower cost in 1990s. management part can be taken care by the host.
The cloud symbol was first introduced to delimit the Hence this type of solution is charged higher
function or area between the provider and the users. where the enterprise is charged for the hosting,
Later, the cloud extends this boundary and covers the server, security, power, cooling and other
computing services and network infrastructure. In 1997 hardware. Hence, only a large company or an
Chellappa first used the term “Cloud Computing” on enterprise can afford this type of offering.
the INFORMS. 4. Strict control and compliance can be achieved
The main concerned of cloud computing is the with a private cloud and the risks incurred due
sharing and coordinated use of diverse resources in to third parties can be greatly reduced.
distributed organizations cloud, which is consisted of 5. Scalability being the core features of Cloud
different, organizations and systems. Through this can be achieved in this type of solution as well.
technology organization large-scale controlled sharing
B. Public Cloud
and interoperation among resources that are dispersedly
owned and managed. We can build scalable cloud 1. The cloud infrastructure is for general public
infrastructure in multiple data centers using dedicated or a large industry group
and cloud servers with elasticity load balancing and 2. In public cloud website, files, applications,
cloud storage with the total control through automation data etc. are stored over the infrastructure
and self service. With the use of cloud organization can offered by a hosting provider. The host would
add consistency reliability efficiency load balancing charge from customer for using their resources
features to their infrastructure. No doubt security is a which may usually be in two modes i.e. fixed
monthly scheme similar to all other plans or II. BENEFIT OF HYBRID HOSTING
pay-per-use model. [5]. OVER CONVENTIONAL HOSTING
3. In public cloud customer not required to pay
an upfront cost for the hardware used. A. Meet Customer Demand without Capital
The essential software’s are usually Expenditure
pre-installed and are included with the package This is the one of the most important benefit of the
4. With a managed Public Cloud, users tend to hybrid cloud computing environment as capability and
enjoy multiple benefits such as on-demand resources can be paid according to the usage of the
technical support, expert guidance, security of resources and facilities. Therefore, capital costs are
our data, flexibility to upgrade or downgrade. reduced and replaced by manageable, scalable operating
5. Customer gets to avail the incomparable expenses. Support and maintenance cost reduced, as the
features such as scalability and reliability. support and maintenance services provided by the cloud
Few hosting providers would also offer provider. The cost of highly trained and expensive IT
customer choices between the technologies personnel is also minimized. Ultimately customer can
used over the cloud for example VMware reduce the large investment in computing infrastructure,
Hosting or Hyper Cloud. capital cost, support and maintenance cost, trained IT
personnel through using cloud computing as a tool.
C. Community Cloud
In case of conventional hosting we have to incur
Several customers share the cloud infrastructure capital expenditure on the infrastructure needed for the
and it supports a specific community with the same development as a result support and maintenance cost
concerns (e.g., resources and security requirements, also increased. Also highly trained and expensive IT
policy and compliance considerations, etc). [1]. personnel are increased with the expansion of the
infrastructure. [6].
D. Hybrid Cloud
B. Custom Networking or Appliance Solutions
1. Public and Private Cloud solves the
requirement for both Internal as well as Hybrid hosting provides a Colocate a security,
External solutions. But what if someone mail, or transcoding appliance, and connects to the
wishes to use both the types to suffice the cloud via a secure private network. With the use of
requirements of a business? This can be hybrid hosting we can save our cost which we have to
achieved by opting for a Hybrid Cloud wherein incur in case of conventional hosting.
some part of the information can be kept on a C. Low Risk to our Business Application & Data
private Cloud while information with the least
priority can be maintained over the public With the use of hybrid hosting user can make their
cloud. [6]. data and application which are running online more
2. Most Hybrid Cloud solutions make use of reliable and accessible as and when needed. The hybrid
virtual front end servers and physical database environment provides a secure way to today’s
servers. organization that are running online application and
3. Virtualization techniques can be implemented earning because of their application. Data is also very
over the existing infrastructure for reducing the important for the organization. Hybrid hosting provides
costs and increasing the reliability.[8] an environment in which our data is secure from the
4. The designing, implementation and risk of the data lost in case of failure of dedicated
management of the virtual network over the server. Similarly applications will continuously working
existing infrastructure needs to be taken care even in case of failure of dedicated server. Dynamic
by the enterprise, while having a public cloud load balancing is also very important feature of hybrid
from a provider can reduce your costs over hosting.
employment of trained labor. In case of conventional hosting if the dedicated
When choosing the best type of solution, we must server is down due to any reason the applications rely
identify the requirement of the extent of control we on that particular server are stop. Data might not
wish to have over our data. If we wish to keep recover in case of dangerous fault with the server such
everything on a private network, we may opt for a as hard disk crash of the server. These risk covers
Private Cloud. But if we wish to keep your expenses completely with the use of hybrid hosting.
low, choosing a Public Cloud can best suffice our
D. Hardware Customization and Compliance
requirements. Moreover, if we are too low on budget
and have a limited size of data yet wish to host our files A major benefit of cloud computing is the
over a cloud, a Shared Cloud can be a perfect choice to flexibility that is provided by the following:
suffice the requirements. 1. Freedom from concerns about updating
servers.
2. Freedom from having to install software Scalability is also possible with the conventional
patches for updating. hosting with the limited scope. [4]
3. Automated provisioning of new services and
technologies. G. Flexibility
4. Acquiring increased resources on an as-needed Hybrid hosting offers a great deal of flexibility
basis. which is essential for today’s business world as
5. Ability to focus on innovation instead of information technology needs are always changing and
maintenance details. evolving which requires changes in infrastructure. A
6. Device independence. hybrid hosting plan is an effective solution as it offers a
Cloud service provide offer a choice among a simple streamlined data migration process that quickly
number of computing and storage resource delivers quality ensuring that there is no loss of
configurations at different capabilities and costs, and productivity.
the cloud customer will have to adjust his or her When using a hybrid solution, users can use the
requirements to fit one of those models. Resiliency can cloud platform to develop and test which makes the
be achieved with the availability of multiple redundant process of development much easier and more efficient.
resources and locations. Resources are available on-demand and on a pay-per-
Such choices and flexibilities are not possible in use basis. In addition, if there are compliance issues, a
case of conventional hosting. dedicated server, or colocation environment can be
created to store data. And if there is a disaster within the
E. Disaster Recovery
infrastructure, information will be replicated to a
Disaster recovery in case of conventional hosting is secondary data center to guarantee the security of data,
not powerful as in case of hybrid hosting. In case of programs, and applications. These flexibilities are not
Hybrid Hosting there are compliance issues, a dedicated achievable with the conventional hosting. [10]
server, or colocation environment can be created to
store data. And if certain conditions arise that leads to H. Better Data Security with Backups
disaster the data will be replicated to a secondary data Security measures adopted by cloud vendors are
center. The ultimate user of the information not very strict. These vendors serve a number of clients and
acknowledges that there is some problem with the are capable of dealing with frequent security upgrades
original infrastructure. Disaster recovery and business and patches needed for the cloud security. In most
continuity planning techniques are employed by the cases, backups of self-hosted solutions are performed
cloud provider in comparison to the local computer once a day. Data could be lost if a piece of hardware
environment of an organization. Cloud provider fails and the most recent backup is close to 24 hours
employs more dynamic and skilled computer personnel old. Cloud computing providers ensures that very little
who are master in the disaster recovery process. As the data will be lost if a hardware failure occurs because
cloud serves to many client at the same time and backups are performed frequently. In some cases,
resources are used by so many users at the same time backups are performed in real-time. [8]
hence cloud provider adopt more advance technique
and skills to handle disaster recovery.[3] I. Quicker Deployment
maintenance cost. In contrast, with cloud computing Load balancing is also an effective feature of the hybrid
services such as take care of all maintenance cloud. The data can be easily recovered even if our
requirements, upgrades and patches, are done by the dedicated application and DBMS server is down due to
cloud service provider. any reason. In short we can say hybrid hosting has so
many benefits over the conventional hosting.
K. Access to Data is Convenient and Continuous
Every business needs continuous, convenient ACKNOWLEDGMENT
access to its data. With cloud computing, no matter The authors would like to express their cordial
where a user is located or what time it is, if there's an thanks to Dr. Kazim Naqvi (System Analyst at Jamia
Internet connection the user will have instant access to Millia Islamia, University, India) & Mr. Mujahid
the company's data. [10] Mashir CEO Bitinbit Technologies (India) for their
valuable advice and support to design architecture for
III. CONCLUSION Hybrid Hosting.
Cloud computing helps the modern organization to
run their application with minimum chance of failure. REFERENCES
With the help of cloud customer can deploy their [1] IJCSNS International Journal of Computer Science and Network
application in very short time frame but there are risk Security, VOL.11 No.5, May 2011.
associated with it also. With the use of Hybrid Hosting [2] NIST Special Publication 800-27 Rev A.
[3] IJCSNS International Journal of Computer 224 Science and
in a cloud environment modern organizations take Network Security, VOL.11 No.1, May 2011.
benefits of both Private and Public cloud. For non stop [4] International Journal of Computer Networks (IJCN), Volume (3)
working of the application over the internet hybrid : Issue (5) : 2011.
plays a very important role. With the use of Hybrid [5] International Journal of Computer Science and Security (IJCSS),
Volume (5) : Issue (4) : 2011.
Hosting we can secure our data and application even if [6] www.gogrid.com/cloud-hosting/hybrid-hosting.php
our dedicated server is down due to any reason. [7] Journal of Service Science-2010 Volunte 3, Number 1.
Infrastructure cost and maintenance cost also reduced [8] Communications of the ac m | April 2010 | vol. 53 | no. 4.
due Hybrid hosting makes the availability criteria of the [9] P. Mell. T. Rrance “ The NIST Definition of Cloud Computing
(Draft)”, NIST Jan 2011.
data and the application wider. Performance also [10] Ronald L. Krutz, Russell Dean Vines-Cloud Security A
increases with the use of hybrid hosting as the data is Comprehensive Guide to Secure Cloud Computing 2010.
stored over the dedicated servers and the cloud server.
Abstract—Since software’s are developed for every hospital management is also considered very important
small or big organization. Also with software development aspect which should be handled by hospital
tasks becoming easy and simple, software is now being
administration, hospital management includes
used widely in every field, and with its usage people are
getting benefits too. In this paper we are comparing three management of all departments in hospital. In this paper
different projects where RUP (Rational Unified Process) is the department discussed is outpatient department. In
being used as a software development process. Case 1 any hospital the outpatient facilities provide the linkage
depicts a project in which a software is developed for of the hospital with the patients/ community. Outpatient
hospital (Outpatient Department). Outpatient department service has wide scope in providing medical services on
is the first place where patient steps in for treatment, and scheduled and unscheduled basis (Francis 1991).
then after consulting to patient doctor decides whether
patient should be admitted to hospital or not. Case 2
Outpatient Department (OPD) is defined as a part of the
presents a project which is developed for tracking missing hospital with allotted physical, medical facilities, their
cell phones. In this project cells are traced using their staff in sufficient numbers with regular scheduled hours
IMEI number and Case 3 discusses a project which is to provide care for patients who are not registered as
developed for a businessman consisting of a website, inpatient (Ross 1985). OPD is definitely a superb
developed for purchasing fresh vegetable and fruits online service, facilitating doctors and patients. Improving it
and does home delivery. For developing these software
by adding some advance technology or for making it
RUP (Rational Unified Process) is being used as a software
development process due to its core principles i.e. work more smoothly software is developed which will
architecture centric, and use case driven. Provides model help saving patient’s time and providing good service to
for user documentation and is an iterative and patient by facilitating doctor sitting inside to keep track
incremental process model. RUP defines many artifacts of patient queue waiting outside. In present scenario
and roles which are very useful while developing software. doctor’s were unable to find the exact number of patient
In our team no one had a thorough knowledge of using
waiting in queue for their turn. This software is a
RUP, but some had read literature and others have a
rough knowledge of RUP, gained while doing some desktop application in which the doctor’s assistant
projects where RUP was used. In all these three cases RUP enters the name of the patient who has came to visit him
was used effectively for software development. In this and doctor has the list of people who are waiting for
paper all the three cases are compared and discussed how their turn, when the patient has visited doctor its name
RUP is tailored in all these three projects using UML is removed from the list and the list refreshes after
diagrams. All the three projects are developed for every 5 minutes to have recent entry. Case 2 depicts a
different clients and for different purposes, it is been
shown in this paper how best, efficiently RUP can be used project which is developed for a detective firm whose
to develop designs of different projects and how it is name is also not disclosed. In this project a software is
tailored to achieve the projects requirements of developers developed which is used for tracking cell phones which
as well as customer for communication and other legal are lost with the help of the lost cell phones IMEI
requirements. number which is unique for each mobile. In this project
Keywords: RUP (Rational Unified Process); OPD a software is developed in which details of the owner is
(Outpatient Department); UML (Unified Modeling noted and their IMEI number is also noted and with the
Language); IMEI (International Mobile Equipment help of IMEI number mobiles are traced and if found
Identity). then with the help of details noted cells are refunded to
their owners. Case 3 reveals a project which is a web
I. INTRODUCTION application developed for a business man. This website
In this paper three different projects are discussed. is developed for online shopping of fresh fruits and
Case 1 is developed for OPD in hospital. The hospital is vegetables. With the help of this website, customers can
very popular and contains 150 beds, due to some purchase fresh vegetables and fruits online and pay
privacy reason its name is not disclosed. Nowadays payment online and then the stuff purchased is
delivered to their respective homes. The process model
used for developing all the three projects is RUP. RUP engineering [2].The UML diagrams used are described
is a disciplined, iterative approach used to assign task in next section.
and responsibilities to software development team. Uses
B. Benefits of This Project
UML (Unified Modeling Language) for visualizing
model. Different UML diagrams are required in all this 1) Benefits to doctor
cases to explain the projects to their entirety. The RUP
is an adaptable process framework that is architecture- This project really proved a boon for doctor, they
centric and risk-driven. In RUP, software engineering can now well manage their patients within the time
limit decided by them, as by knowing the number of
processes are organized into phases and iterations [1].
patients waiting for their turn. Doctor can decide his
A. Case 1 schedule accordingly and by knowing the name of
patient doctor can manage its time by knowing how
Outpatient department is nowadays present in much time the patient needs if the patient is a old
almost every hospital, it is a very good initiative by patient i.e. the patient that comes for regular checkup
hospitals as it is also very beneficial to patient also, and if that patient need to refer to a surgeon or to any
OPD allow patients to take treatment by good doctors other department for any test than doctor can send the
without being admitting in hospital. OPD optimizes the patient to other department without wasting much of
resources and is an outstanding service for effective patient’s time in waiting. Doctors get opportunity to
health care as patient get treated well without paying tackle different cases which increases their experience
any hospital charges for bed and other services, doctor and helps in diversifying their practice.
get full support from hospital administration and inner
department, they can send patient to surgeon if 2) Benefits to patients
necessary which are available inside hospital. Patient
It was also beneficial for patient as if the queue is
also need not to stay hospital for days or weeks, they
long they can ask doctor assistant will their number will
can go to home after the doctor consultation and get
come today or not and by seeing the name of patients
proper care at their homes as per doctor advices.
waiting doctor can give an estimation that after how
For OPD we developed an desktop application for
much time their number will come. This would
doctors sitting at OPD for knowing the number of
definitely save patient time. It also saves patient money
patients waiting outside for their turn, so that doctors
by providing good infrastructure and services at low
can arrange their schedule accordingly and if there are
cost, saves hospital bed charges and other hospital
many patients waiting outside for treatment they can
charges. It also saves patient from visiting to knapsack
ask for some more doctor to come to OPD for handling
doctor and ensures patients to be treated from
patients from hospital administration, so that every
authenticate doctors.
patient can be consulted without overloading a single
doctor and patient also do not have to waste time by 3) Benefits to hospital
sitting in queue and waiting for their turn. This software
helps patient saving their time by doctor getting the list This project is also very beneficial for hospital as
of patient waiting and by doctor taking appropriate their resources(beds and man power) are optimized.
action. Outpatient Department in any hospital is considered to
The organization which developed this software is be shop window of the hospital [3, 4]. Now a day, the
a medium sized organization consisting of 50 people. patients are looking for hassle free and quick services in
The organization name is not disclosed due to some this fast growing world. This is only possible with
reason, organization has developed many projects for optimum utility of the resources through multitasking in
public and private sector. Mainly project developed a single window system in the OPD for better services
belong to public sector like for police and hospital. This [5,6]. Hospital can earn well using OPD as patient feel
project was developed by developing team consisting of easy to go and get treated at OPD at low cost and by
15 people including project leader, tester and good doctors without being admitted to hospital.
developers. Technology being used was PHP. Software C. UML Diagram Used
process model used was RUP. RUP is a software
engineering process used to develop software. It is an RUP defines UML diagrams for better
iterative software development process framework understanding of project by developers and
based on UML. No RUP mentor was appointed for stakeholders. Unified Modeling Language (UML) is a
training. Some team members has RUP knowledge due standardized general-purpose modeling language in the
to reading literature and some has worked on project field of object-oriented software engineering [7]. Since
using RUP before. UML (Unified Modeling Language) UML is not a methodology, it does not require any
diagrams were used for creating or visualizing models. formal work products. Yet it does provide several types
UML (Unified Modeling Language) is a standardized of diagrams that, when used within a given
general purpose modeling language in field of software methodology, increase the ease of understanding an
application under development [8]. In this project UML diagrams to visualize the workflow of a business use
diagrams are used for easy understanding of project case. A complete workflow description will have a
goals. The most useful, standard UML diagrams are: basic flow, and one or several alternative flows. This
use case diagram, class diagram, sequence diagram, workflow has a structure that we can define textually,
statechart diagram, activity diagram, component using informal if, if-then-else, or do-until statements of
diagram, and deployment diagram. But in this project various kinds. For a simple workflow with a simple
we have used only use case and activity diagrams. Out structure, such textual definitions may be quite
of 9 diagrams mentioned above only combination of use sufficient, but in the case of more complex structures,
case diagram and activity diagram are required to activity diagrams help to clarify and make more
explain the scope and functioning of our software. both apparent what the workflow is [12]. Activity diagram is
the diagrams are discussed below which will explain the used to show the flow of the events defined in project. It
software to its entirety. captures the dynamic behavior of the project.
4) Use case diagram
Use case diagrams are behavior diagrams used to
describe a set of actions (use cases) that some system or
systems (subject) should or can perform in
collaboration with one or more external users of the
system (actors). Each use case should provide some
observable and valuable result to the actors or other
stakeholders of the system.
Use case diagrams are used to specify:
• (external) requirements on a subject, required
usages of a system-to capture what a system
under construction is supposed to do;
• The functionality offered by a subject–what
system can do;
• Requirements the specified subject poses on its
environment-by defining how environment
should interact with the subject so that it will
be able to perform its services. [9]
Actors–Doctor’s Assistant, Doctor
D. Case 2
In this a software is developed which is developed
for a private detective agency for tracking the lost cell
phones . This software is developed by a team
consisting of 10 peoples including tester, coder and
project manager. this software was developed using C#
technology and for developing this project RUP was
Use case diagram also depicting scope of our used as s software process model.
project. The IMEI (International Mobile Equipment
Activity Diagram–An activity diagram is a simple Identity) number of a mobile phone is a 15 digit number
and intuitive illustration of what happens in a workflow, unique to a mobile handset. It is usually printed on the
what activities can be done in parallel, and whether compliance plate of the handset usually under the
there are alternative paths through the workflow. battery. IMEI number can help in tracking or blocking
Activity diagrams as defined in the Unified Modeling mobile phone if it is lost. When a phone is switched on,
Language [10] are derived from various techniques to the IMEI number is transmitted and checked against a
visually illustrate workflows. In the Rational Unified database in the network's EIR (Equipment Identity
Process [11] we talk about how you can use activity
Register). The EIR has three categories namely the the website and their rates are tagged below the
"white lists", "grey lists" and "black lists". The operator respective items. Item which are out of stock is also
can block a handset by putting it in black list. Also by mentioned in the website, so the buyers can purchase
using Global Positioning System an operator can track the stuff they needed without coming to market sitting
the handset user also[13]. at their home. They can pay amount using their bank
In this software details of the person whose cell cards and the purchased stuff will be delivered to their
phone is noted like his name, address, alternate number, respective homes.
city, state, model number and IMEI number. Then the This software was developed by a team of 15
IMEI number is put in black list for blocking the members including coder, tester and project manager.
number and as soon as the cell phone is traced, it is the technology used for developing this website was
returned back to user through his/her details entered. Zend framework, it is a PHP framework. The software
is developed in 20 days and after testing it is installed to
E. Benefits of this Project user machine. Database is also created for noting the
1) Benefits to user customer details. Database used was MySql. RUP was
used to develop this software.
Users are benefited by this software very much as
cell phone is valuable thing and apart from mobile cost H. Benefits of this Project
it also contain valuable of information of person whose
2) Benefit to customer
cell phone is lost. With the help of this software they
can get back their cell phones. Life nowadays is very busy, so people does not get
time to go to market everyday and buy fresh vegetable
2) Benefit to agency and fruits, so they often stock vegetable and fruits for a
With the help of this software, agency work has week. But it may cause adverse effect on health. So for
become smoother as they don’t have to keep manual customer comfort this website will deliver fresh fruit
record of their customer. They can now keep track of and vegetable to their home without wasting their time
people who have lost their cell phones. With the help of to go to market and purchase the daily required food
this they can keep record of case solved and case stuff. They can now sitting at home or from their office
pending. just select the food stuff needed, pay the amount and
their ordered stuff will be delivered to their home.
F. UML Diagrams Used
3) Benefit to trader
In this project the UML diagram used for
explaining this project to its entirety and for reducing The trader can sell his food stuff now through
the development time by enabling developers to website too. It will definitely increase his sales. By
understand the project completely is use case diagram. knowing which stuff is ordered more he can order that
stuff in bulk amount. It will save his investment money
G. Use Case Diagram also as by knowing his sales per day through his
website.
I. UML diagrams used
4) Use case diagram
TRACK V
ENERGY AND SOCIAL ENVIRONMENT
ISBN: 978-93-82880-20-2 3
Proceedings of National Conference on Trends in Signal Processing & Communication (TSPC’13), 12th–14th April 2013
Abstract—Renewable energy resources like wind turbines [4]. These grid codes established similar
hydro/wind power plants play an important role in requirements on usual generating plants to wind power
upcoming future of power grids. Thereby, a full flexibility converters, i.e. supporting the network stability in
of utilizing the grid capabilities in terms of active and normal operation and during faulty conditions. During
reactive power production in generating mode as well as
in motoring mode gains in significance. In order to
normal operation this is translated into the capability of
provide a continuous power delivery during voltage dips, frequency regulation through active power control and
control of active and reactive powers generated is of voltage regulation through reactive power control.
desirable for the purpose of grid support. This paper The work presented in this paper is concentrated on this
reviews available technique for power control of doubly latter theme and proposes various control strategies for
fed induction machine. The control strategies such as the power flow in a doubly-fed induction generator
vector control and direct torque control techniques are (DFIG) drive aiming the reduction of the electrical
analyzed and detailed in order to achieve the best overall power losses. In order to enable variable speed
performance. operations with a lower rated power converter, doubly-
Keywords: Active and reactive powers, Asynchronous fed induction generator (DFIG) can be used as shown
machine, Doubly fed induction machine, direct torque on Fig. 1. The stator is directly connected to the grid
control, frame transformation, vector control. and the rotor is fed to magnetize the machine with the
help of two back-to-back PWM converters.
I. INTRODUCTION
Due to the increasing concern about environmental
pollution and a possible energy shortage, the renewable
energy systems and generation have attracted great
interests in recent years [1]-[2].Countries where this
development first took place like Germany and
Denmark, started using wind as well as hydro energy as
a major consent for power generation. The considerable
amount of energy generated and delivered to the
interconnected electrical network depends
instantaneously on site weather conditions. For wind
power generation, experience tells that wide-spread
wind generation results in more stable average
generated powers [3], the dependency on the actual
wind conditions is still a risk factor for the balance Fig. 1: Wind Turbine with Doubly Fed Induction Machine
between energy offer and demand as a basic condition In this paper, the control of electrical power
for reliable operation of the electrical power system. On exchanged between the stator of the DFIG and the
the other hand the increased machine powers are power network by controlling independently the torque
accompanied by a nearly proportional increase on the (consequently the active power) and the reactive power
losses. Electrical machines in the MW-class present is presented [5]. Several investigations have been
very high efficiency values reaching saturation with the developed in this direction using cycloconverters as
state-of-the-art materials and construction technology. converters and classical proportional-integral regulators
Therefore only slight augmentation on efficiency is to [6]-[8]. In our case, after modelling the DFIG and
be expected with the progressing generator rated power. choosing the appropriate d-q reference frame, active
In order to avoid network stability problems power and reactive powers are controlled using respectively
companies launched guidelines for net connection of Integral-Proportional (PI). Their performances are
compared in terms of reference tracking, sensitivity to The mechanical power and the stator electrical
perturbations and robustness against machine's power output are:
parameters variations. Pm =Tm ωr
Ps =Tem ωs
II. MODELING OF THE WIND TURBINE For a lossless generator the mechanical equation is:
DOUBLY-FED INDUCTION GENERATOR
The wind turbine and the doubly-fed induction = −
generator are shown in Fig. 1. The AC/DC/AC
In steady-state for a lossless generator we have:
converter is divided into two components: the rotor side
converter Crotor and the grid-side converter Cgrid. = = +
Crotor and Cgrid are voltage-Sourced Converters that from where it follows:
use forced-commutated power electronic devices
(IGBTs) to synthesize an AC voltage from a DC − = − = −
voltage source. A capacitor connected on the DC side Where = is defined as the slip of the
acts as the DC voltage source. A coupling inductor L is generator.
used to connect Cgrid to the grid. The three-phase rotor Generally the absolute value of slip is much lower
winding is connected to Crotor by slip rings and than 1 and consequently Pr is only a fraction of Ps.
brushes and the three-phase stator winding is directly Since Tm is positive for power generation and since ωs
connected to the grid. The power captured by the wind is positive and constant for a constant frequency grid
turbine is converted into electrical power by the voltage, the sign of Pr is a function of the slip sign. Pr is
induction generator and it is transmitted to the grid by positive for negative slip (super-synchronous speed)
the stator and the rotor windings. The control system and it is negative for positive slip (sub-synchronous
generates the pitch angle command and the voltage speed). For supersynchronous speed operation Pr is
command signals Vr and Vgc for Crotor and Cgrid transmitted to DC bus capacitor and tends to rise the
respectively in order to control the power of the wind DC voltage. For sub-synchronous speed operation, Pr is
turbine, the DC bus voltage and the voltage at the grid taken out of DC bus capacitor and tends to decrease the
terminals.An average model of the AC/ DC/ AC DC voltage. Cgrid is used to generate or absorb the
converter is used for real-time simulation. In the power Pgc in order to keep the DC voltage constant. In
average model power electronic devices are replaced by steady-state for a lossless AC/ DC/ AC converter Pgc is
controlled voltage sources. Vr and Vgc are the control equal to Pr and the speed of the wind turbine is
signals for these sources. The DC bus is simulated by a determined by the power Pr absorbed or generated by
controlled current source feeding the DC capacitor. The Crotor. The phase-sequence of the AC voltage
current source is computed on the basis of instantaneous generated by Crotor is positive for sub-synchronous
power conservation principle: the power that flows speed and negative for super-synchronous speed. The
inside the two AC-sides of the converter is equal to the frequency of this voltage is equal to the product of the
power absorbed by the DC capacitor [9]. The power grid frequency and the absolute value of the slip. Crotor
flow illustrated in Fig. 2 is used to describe the and Cgrid have the capability for generating or
operating principle. absorbing reactive power and could be used to control
the reactive power or the voltage at the grid terminals.
Fig. 3: Third Step of the DFIG Connection Vector control can also possible with air gap flux
or stator flux orientation, but at the cost of a coupling
IV. CONTROL SCHEME OF DFIG effect that demands decoupling compensation. Stator
flux oriented direct vector control has the advantage
Many variable schemes use a power converter
that flux vector estimation accuracy is estimated by the
between the induction generator stator and the supply
stator resistance Rs variation only. In this control we
thus necessitating that the converter volt ampere rating
developed a strategy for stator flux oriented vector
matches that of the generator output. Additionally if the
rotor current is controlled using vector methods then control by using the equations derived from d-q
fast decoupled control of stator active and reactive equivalent circuits. If the stator flux is oriented on the d
power will result. The opportunity exists to produce axis, then the flux q-axis component Ψqs=0. In the
power at unity power factor without the need for stator flux orientation, the active and reactive powers
capacitor compensation. Schemes exists which operate are decoupled and can be controlled via the rotor
using controlled rectifiers [11] but the resulting current currents. By means of the direct current, we can control
waveforms have relatively high harmonic content. The the active power while the reactive power can be
use of a twin PWM bridge inverter allows the inverter controlled via the quadrature current.
network connection to contribute to high levels of
power quality through the use of a sinusoidal rectifier. C. Establishment of Control System
In this paper we examine the vector control scheme The control scheme for the doubly fed induction
for DFIG. generator consists of two cascaded control loops. The
outer loop is concerned with the control of active and
A. Mathematical Model of the DFIG
reactive power so that the power factor value
For a doubly-fed induction machine, the Park determined by the electric energy distribution company
transformation’s application to the traditional a,b,c is complied with as accurately as possible.
model allows to write a dynamic model in a d-q
reference frame as follows:
= + −
= + +
= + −
vector relative to the stator or supply voltage vector and Fig. 7 shows the schematic control of a Grid side
this is done through rotor current control as shown converter control.
in Fig. 4.
Where Kp and Ki are inner control loop PI controller
parameters.
On the other hand, the inner control loop task
consists in controlling independently the rotor current
direct idr and quadrature iqr components expressed
according to the reference frame fixed to the stator flux
linkages space phasor. In order to implement this inner
control loop, the stator flux oriented vector controls
based on two identical PI controllers are used [12] as
shown in Fig 5.
REFERENCES
[1] Fadaeinedjad R,Moallem M and Moschopoulos G, “Simulation
Fig. 6: Control Structure of Rotor Side Converte of a Wind Turbine With Doubly Fed Induction Generator by
FAST and Simulink”[J].IEEE Transactions on Energy
2) Grid side converter control Conversion, Vol.23, No.2,pp.690–700,2008.
[2] Gomis Bellmunt 0,Junyent Ferre A,Sumper A,et al, “Ride-
The Grid Side Converter (GSC) ensures the Through Control of a Doubly Fed Induction Generator Under
regulation of the DC bus voltage and adjusts the power Unbalanced Voltage Sags”[J]. IEEE Transactions on Energy
factor on the grid side. The GSC is a bidirectional Conversion, Vol.23, No.4, pp. 1036–1045, 2008.
[3] S. Heier, Wind Energy Conversion Systems, 2nd Ed. New York:
converter which operates as a rectifier when the slip (S) John Wiley & Sons, 1998.
is positive (sub synchronous mode) and as an inverter [4] Grid Code High and Extra High Voltage, 2nd ed. E. On Netz
when the slip is negative (super synchronous mode). GmbH, 2006.
[5] L. Xu, W. Cheng, "Torque and reactive power control of a [9] Branislav Dosijanoski, "Simulation of Doubly-Fed Induction
doubly-fed induction machine by position sensorless scheme.", Generator in a Wind Turbine”, XI International PhD Workshop
IEEE Transactions on Industry Applications, Vol. 31, No.3, OWD 2009, 17–20 October 2009. pp. 378–380.
pp 636–642, May/June 1995. [10] L. Morel, A. Mirzaian, J.M. Kauffmann: "Field oriented control
[6] M. Yamamoto, O. Motoyoshi, "Active and reactive power for double fed induction machine: simulation and experimental
control for doubly-fed wound rotor induction generator.", IEEE results.” ELECTRIMACS’96. Vol. 2, September 17-18-19,
Transactions on Power Electronics, Vol. 6, No. 4, Oct. 1991, 1996, pp 391–396.
pp 624–629. [11] Li, R., Wallace, A., and Spee, R. 1995.”Determination of
[7] M. B. Rifai, T. H. Ortmeyer, "Dynamic analysis of a doubly fed converter control algorithms for brushless doubly-fed induction
generator in power system applications.", Electric Machines and motor drives using Floquet and Lyapunov techniques”, IEEE
Power Systems, Vol. 21, pp 141–150, 1993. Trans, Power Electronics, 10(1), Jan.78–84.
[8] B. Hopfensperger, D.J. Atkinson, R.A. Lakin, "Stator-flux- [12] Srinath Vanukuru and Sateesh Sukhavasi,” Active & Reactive
oriented control of a doubly fed induction machine with and Power Control Of A Doubly Fed Induction Generator Driven By
without position encoder.” IEEE Proc.-Electrical Power Appl., A Wind Turbine”, International Journal of Power System
Vol. 147, No. 4, July 2000, pp 241–250. Operation and Energy Management, ISSN (PRINT):
2231–4407, Volume-1, Issue-2, 2011, pp.86–88.
Abstract—The purpose of this paper is to explain the smart grid technology more than any other country,
importance of Smart Grid Deployment in INDIA. The which if conservatively can reduce the losses to 20%
paper throws the light on various problems and challenges then it will provide enormous boost to growth.
related to electricity grids that are faced in INDIA and
the solution to manage those problems by adopting the
vision of a ‘Smart Grid.’ The essence of this vision is ‘a II. SMART GRID
fully-automated power delivery network that can ensure a The essence of smart grid is a promise of
two-way flow of electricity and information between the transforming our current electricity delivery system into
power grids and appliances and all points in between’.
a more efficient and secure system that is able to better
The further explanation is on the Ongoing and Future
Projects on Smart grid in INDIA. Finally, the paper integrate variable supply resources while giving the
concludes with the suggestions and proposals for the end-user, the customer, greater autonomy in his/her
Regional Approach to Smart Grid Development that energy consumption.
explains the demand for Smart Grid capability within According to the IntelliGrid Architecture Report,
INDIA. the Electric Power Research Institute (EPRI) defines the
Smart Grid as:
I. INTRODUCTION • A power system made up of numerous
automated transmission and distribution
The Republic of India is the seventh-largest
(T&D) systems, all operating in a coordinated,
country by geographical area and the second by
efficient and reliable manner.
population. Like the United States, it has national,
regional and local governments, including 28 states and • A power system that handles emergency
7 union territories. It borders Pakistan to the west; conditions with ‘self-healing’ actions and is
China, Nepal, and Bhutan to the northeast; and responsive to energy-market and utility needs
Bangladesh and Myanmar to the east. India is one of the • A power system that serves millions of
fastest growing economies in the world cloaking an customers and has an intelligent
average annual growth rate of 7%+ for last 6 years. communications infrastructure enabling the
Simultaneously, the current Indian population at 1.1 timely, secure, and adaptable information flow
Billion is growing at the rate of 1.4 % annually, and it is needed to provide power to the evolving digital
estimated that India will be the largest populated economy.
country by 2050 AD. Energy is the most critical issue From this definition, we can conclude that the
for human development and growth now. At present, Smart Grid must be:
India is unable to meet the existing need of energy to • Predictive (operationally and functionally) to
her inhabitants. Though electricity is available to 80% preclude emergencies.
of the population, it is irregular and insufficient. The • Self-healing to correct/bypass predicted/
power interruption is a chronic problem with on average detected problems.
17 power-cuts per month to manufacturing sector alone. • Interactive with consumers and markets.
The total installed capacity is 210951.72 MW as of • Optimizable to make the best use of resources
December 2012 data published by Ministry of power on • Distributed in nature with both assets and
its website. However, it is way shorter than the present information.
need and requires enormous investment to meet the • Transformational to turn data into information.
existing and future need of the growing population and • Secure from threats and hazards.
economy. This indicates that India is in dire need of
Energy. New power plants are being built and more will III. GRID PROBLEMS IN INDIA
require however, it want suffice the growing need.
Indian transmission and distribution losses are amongst As is the case in most of the world, the Indian
the highest in the world amounting 26% of the national grid was not designed for high-capacity, long-
production. If non-technical losses like theft are distance power transfer. India needs to interconnect
included into total, it is around 50%, hence the total regional grids. Although coal and hydro-electric
effective distribution is only 105475.86 MW out of potential has peaked in many parts of India, there are
210951.72 MW Installed capacity. Thus, India needs still several regions with excess capacity. Large wind
potential and increasing wind capacity in the south and begin a drive to promote energy efficiency in
west also create a need for transmission infrastructure. the country.
Unfortunately, regions are generally sectionalized, with 3. The Bureau of Energy Efficiency has also
some asynchronous or HVDC links allowing for come out with an Action Plan that includes
minimal power transfer. India’s transmission grid only various schemes to promote saving of
reaches 80% of its population. According to its Ministry electricity.
of Power, India’s transmission and distribution losses 4. The Government of India is also working to
are among the highest in the world, averaging 26% of develop advanced technologies of renewable
total electricity production, with some states as high as energy such as: geothermal energy, ocean
62%. When non-technical losses such as energy theft energy, fuel cells, bio fuels and tidal energy to
are included in the total, average losses are as high as satisfy future energy requirements.
50%. The financial loss has been estimated at 1.5% of 5. Every year, National Energy Conservation
the national GDP, and is growing steadily. India’s Award is conferred to companies who
power sector is still largely dominated by state utilities. contribute for conservation of energy.
Despite several attempted partnerships with foreign 6. Under National Program on Improved Chulha
investors, few projects have actually been implemented. (NPIC) (Smokeless cook-stoves), Government
This lack of foreign investment limits utilities’ ability to of India provides smokeless and improved
raise needed capital for basic infrastructure. wood cook-stoves (both portable & fixed) and
India’s grid is in need of major improvements. This provides Central Subsidy.
neglect has accumulated in a variety of system failures: 7. Besides, day to day awareness campaign,
• Poorly planned distribution networks. painting competition among school children on
• Overloading of system components. energy saving day are organized.
• Lack of reactive power support and regulation 8. The Ministry of New and Renewable Energy
services. (MNRE) has initiated certain comprehensive
• Low metering efficiency and bill collection. programs for the development and utilization
• Power theft. of renewable energy.
known elsewhere as ‘curbstone readings’ or ‘shade tree’ B. India Smart Grid Task Force
readings, or even deliberate errors, which are thought to The Smart Grid Task Force (ISGTF) was
be significant reasons for losses. constituted under the Chairmanship of Shri Sam
D. Peak Load Management Pitroda, Adviser to Prime Minister, Government of
India on Public Information Infrastructure and
India’s supply shortfalls are expected to persist for Innovation. The Smart Grid Task Force is an inter-
many years. A smart grid would allow more ministerial group and serves as Government’s focal
‘intelligent’ load control, either through direct control point for activities related to ‘Smart Grid’. The main
or economic pricing incentives that are communicated functions of the Smart Grid Task Force to ensure
to customers in a dynamic manner. Such measures awareness, coordination and integration of the diverse
would help mitigate the supply-demand gap. activities related to smart Grid technologies, practices
and services for Smart Grid Research and
E. Renewable Energy Development; Co-ordinate and integrate other relevant
inter governmental activities; Collaborate on
India has supported the implementation of interoperability frame work; Review and validate the
renewable energy. Historically, much of its support was recommendations from India Smart Grid Forum etc.
for wind power, but the newly announced National
Solar Mission and its goal to add 20,000 MW of solar C. Center for Study of Science, Technology,
energy by 2020 should be an accelerant. Spurred by and Policy (CSTEP)
environmental concerns and the desire to tap into all It is a private, non-profit organization with an
available sources of power, this move can also be a objective to carry out research in science, engineering,
smart grid driver. and technology that can be beneficial to country's
economic and human development. CSTEP will serve
F. Technological Sustainable Development as knowledge partner and adviser to Smart Grid India
Perhaps the most intriguing driver for India is the Forum established by Government of India. CSTEP
potential to ‘leapfrog’ into a new future for electricity, published Smart Grid India white paper detailing the
as it did with telecommunications. Also, the ‘smart’ in a Smart Grid in Indian environment.
smart grid is Information and Communication D. Demand Response and Smart Grid Coalition
Technology (ICT), an area of unique capability in India.
It is the trade association for companies that are
VI. VARIOUS SMART GRID GROUPS dealing in Demand Response, Smart Meter, and Smart
Grid technologies. DRSG plays supportive role to
A. India Smart Grid Forum educate policymakers, financial institution, utilities,
media, and other stakeholders that how the smart grid
It is a non-profit, voluntary consortium of public
technologies like demand response and smart meters are
and private stakeholders set up by Government of India.
going to benefit the consumers, the companies, and the
The prime objective of the forum is to accelerate society as a whole. More than 50 companies and
development of Smart Grid technology in India. India associations like Zigbee Alliance and Universal
Smart Grid Forum is also a member organization of Powerline Association are the member of this
Global Smart Grid federation. Ministry of Power, India organization.
is going to be the patron and Power finance Corporation
(PFC) and Rural Electric Corporation (REC) are E. Smart Grid Consumer Coalition
permanent invitee of the forum. Initially, the members It is a non-profit consumer advocacy organization.
will be selected through invitation which will include The objective is to bring together utilities, suppliers,
selected state power utilities, private power utilities, vendors, consumer advocacy groups, and associations
empaneled System Integrators, SCADA Consultants to better understand consumer needs in the Smart Grid
and Implementing Agencies of R-APDRP (Restructured Universe. The group will carry out consumer centric
Accelerated Power Development and Reform research, educate them about Smart Grid, and will share
Programme), power sector PSUs (public sector the best practices with the members.
undertakings), selected educational and research
institutes, NGOs, Central Electricity Authority (CEA), VII. VARIOUS SMART GRID PROJECTS IN INDIA
Central Electricity Regulatory Commission (CERC), A. Uttar Haryana Bijli Vitran Nigam (UHBVN)
Central Power Research Institute (CPRI), Federation of
Indian Chamber of Commerce and Industry (FICCI) This Pilot project is located in Panipat subdivision.
,and NASSCOM. The pilot project covers 30,544 consumers and
distribution system of 531 DTs. The area has around
131.8 MU input energy consumption. The proposed E. Tripura State Electricity Corporation
project area is covered under RAPDRP Scheme for IT Limited (TSECL)
implementation and system strengthening. The
This pilot project, located in Electrical Division
functionality of Peak load management is proposed by
No.1, of Agartala town, covers 46,071 no. of
implementing Automated Metering Infrastructure
consumers. The proposed project area is covered under
(AMI) for Residential Consumers and Industrial
RAPDRP Scheme for IT implementation and system
Consumers.
strengthening. The functionality of Peak load
B. Chamundeshwari Electricity Supply management is proposed by implementing Automated
Corporation, Mysore (CSEC) Metering Infrastructure (AMI) for Residential
Consumers and Industrial Consumers.
This Pilot project located in Additional City Area
Division (ACAD), Mysore, involves 21,824 consumers F. Uttar Gujarat Vij Company Ltd. (UGVCL)
with a good mix of residential, commercial, industrial
This Project proposes covering 20,524 consumers
and agricultural consumers including 512 irrigation
in Naroda and 18,898 agricultural unmetered consumers
pump sets covering over 14 feeders and 473 distribution
in Deesa-II division and accounting for input energy of
transformers and accounting for input energy of 151.89
around 1700MU. The functionalities of Peak load
MU. The functionalities of Peak load management,
management, Outage Management, Power Quality
Outage Management are proposed by implementing
Management are proposed by implementing Automated
Automated Metering Infrastructure (AMI) for
Metering Infrastructure (AMI) for Industrial,
Residential Consumers and Industrial Consumers and
Commercial and Residential Consumers. Some
Integration to Distributed Generation/ Micro Grid
additional functionalities like Load forecasting and
Integration.
Asset Management are also proposed and
C. Electricity Department, Government functionalities of load forecasting, peak power
of Puducherry (PED) management and outage management are also
considered at utility level which will impact all
This project covers 87031 no. of consumers with consumers of utility (i.e. 27 lac consumers) indirectly.
dominant domestic consumers (79%). The area has Renewable energy integration has also been proposed to
around 367 MU input energy consumption. The be carried out at Patan Solar Park and few roof top
proposed project area is also covered under RAPDRP installations at some of the universities.
Scheme for IT implementation and system
strengthening which is likely to be completed in 2013. G. Andhra Pradesh Central Power Distribution
The module of Automated Metering Infrastructure Company Limited (APCPDCL)
(AMI) for Residential Consumers and Industrial
This project is located in Jeedimetla Industrial
Consumers are proposed to be implemented to assist
Area. The Project proposes covering 11,904 consumers.
with consumer issues like event management &
The proposed project area is covered under RAPDRP
prioritizing, billing cycle review and revenue collection
Scheme; DAS, IT and SCADA shall be implemented.
efficiency for Energy auditing and AT&C loss
The functionalities of Peak load management, Power
reduction.
Quality and Outage Management are proposed by
D. Kerala State Electricity Board (KSEB) implementing Automated Metering Infrastructure
(AMI) for Residential Consumers and Industrial
This pilot is proposed for around 25078 LT Consumers.
Industrial consumers of Selected Distribution Section
offices spread over the geographical area of Kerala H. Assam Power Distribution Company Ltd. (ASPCL)
State. The input energy for the total scheme area is
The pilot project covers 15,000 consumers
mentioned as 2108 MUs and for the LT Industrial
involving 90MUs of input energy in Guwahati
consumers is mentioned as 376 MUs. Part of this area is
distribution region. APDCL is in the process IT
covered in RAPDRP scheme. By implementing
Implementation under R-APDRP and SCADA/DMS
Automated Metering Infrastructure (AMI) it is proposed
implementation is also to be taken up shortly. APDCL
to provide quality service, prevent tampering and
has proposed the functionality of Peak Load
unauthorized usage of load, accurate and timely
Management using Industrial and Residential AMI,
metering and billing, avoiding costly field visits of Sub
Integration of Distributed Generation and Outage
Engineers for meter reading, reducing supply
Management system. The utility has envisaged that
restoration time, peak load management through load
Power Quality Monitoring will be a by-product of the
restriction for Remote Disconnection/Reconnection and
deployment.
Time of Day tariff.
I. Maharashtra State Electricity Distribution of AT&C loss reduction and Peak Load Management
Company Limited (MSEDCL) using Automated Metering Infrastructure (AMI) for
Residential and Industrial Consumers
Project proposes covering 25,629 consumers with a
mix of residential, commercial and industrial consumers N. Jaipur Vidyut Vitran Nigam Ltd. (JVVNL)
and input energy of 261.6 MU. The functionality of
Outage management is proposed by implementing Project proposes covering 2646 no. of consumers,
dominated by Industrial consumers (56.46%) and
Automated Metering Infrastructure (AMI) for
around 374.68 MU input energy consumption. Proposed
Residential Consumers and Industrial Consumers. In
project area is also covered under RAPDRP Scheme for
addition MSEDCL has proposed to leverage AMI for
IT implementation and system strengthening. The
Remote connect/disconnect of customers, Monitoring functionality of Peak load management is proposed by
the consumption pattern, Tamper detection, Contract implementing Automated Metering Infrastructure
load monitoring, Load curtailment program i.e. reduced (AMI) for Residential Consumers and Industrial
power supply instead of no power scenario, Time of Consumers.
Use Metering and Dynamic and Real Time Pricing,
Demand forecasting etc. VIII. CONCLUSION
J. Chhattisgarh State Power Distribution ‘Transform the Indian power sector into a secure,
Company Limited (CSPDCL) adaptive, sustainable and digitally enabled ecosystem
by 2027 that provides reliable and quality energy for all
The pilot project includes installing smart meters at
with active participation of stakeholders’, with this
508 H.T. & L.T Industrial Consumer premises as well
vision India is steadily running towards splendid era of
as Automatic Meter Reading (AMR) at 83 DTs. The
Smart Grid. The Smart Grid presents opportunities for
area has around 2140.86 MU input energy
utilities and consumers to benefit from efficient
consumption. The proposed project area is not covered management of energy and advanced equipment and
under RAPDRP Scheme. The functionality of Peak load devices. It offers significant opportunities to wisely
management is proposed by implementing Automated manage the nation's fuel resources by potentially
Metering Infrastructure (AMI) for Industrial reducing the national need for additional generation
Consumers. sources, better integrating renewable and non-
K. Himachal Pradesh State Electricity renewable generation sources into the grid’s operations,
Board (HPSEB) reducing outages and cascading problems, and enabling
consumers to better manage their energy consumption.
The pilot project covers 650 consumers and having According to the US Energy Information
annual input energy of 533 MUs. The functionality of Administration, the world’s electricity generation is
peak load management and outage management is projected to increase by 70 percent from 2012 to 2030.
proposed by implementing Automated Metering So for India the need for grid improvements is
Infrastructure (AMI) for Industrial Consumers, increasing. The country have to face significant
Distribution Automation and Substation Automation challenges in evaluating and shaping utility Smart Grid
and power quality management by deploying Power proposals, since it represents a complex technical
Quality meters at HT consumers challenge. Commission staffs, for the most part, do not
have personnel trained in this area and lack the finance
L. Punjab State Power Corporation Limited (PSPCL) resources needed to hire outside consultants. Existing
The functionality of Outage Management (OM) is electricity grids in India are not sufficient in terms of
proposed to be implemented in the project area for all capacity, efficiency, reliability, security, and
the 85746 consumers and distribution system in area environmental impact to supply the electrical power
using AMI by installing 9000 Smart Meters and by needs of modern societies. Therefore the need of the
Transformer Monitoring. The proposed project area is hour and for modern revolutionizing societies of India
covered under RAPDRP Scheme for SCADA is to adopt Smart Grid Vision.
Implementation and GIS Mapping.
REFERENCES
M. West Bengal State Electricity Distribution [1] Economic Survey of India 2007–08.
Company Limited (WBSEDCL) http://indiabudget.nic.in/es2007-08/esmain.htm.
[2] Power Sector at a Glance "All India",
The pilot project proposes to take up 4 nos. of 11 http://powermin.nic.in/indian_electricity_scenario/introduction.
KV feeders for implementation of Smart Grid covering htm
[3] The Indian Electricity Act, 2003,
4404 consumers. The area has 42 MU input energy http://www.powermin.nic.in/acts_notification/electricity_act200
consumption. The utility has proposed the functionality 3/preliminary.htm
[4] Smart Grid White Paper: The Home Appliance Industry’s [8] Global Smart Energy, ‘The Electricity Economy: New
Principles & Requirements for Achieving a Widely Accepted Opportunities from the Transformation of the Electric Power
Smart Grid, December 2009. Sector’, August 2008.
http://www.aham.org/ht/a/GetDocumentAction/i/44191\ [9] National Energy Technology Laboratory (NETL),’A Systems
[5] Energy Information Administration, ‘Electric Power Annual View of the Modern Grid,’ January 2007.
2007: A Summary’ http://www.eia.doe.gov/ [10] NEMA,’ Defining Intelligence in the Intelligent Electricity
bookshelf/brochures/epa/epa.html, April 2009. Grid,’ Presentation, January 18, 2008.
[6] Detroit Edison, ‘DTE Energy Smart Grid White Paper,’ [11] Smart grids Advisory Council. "Driving Factors in the Move
February 24, 2009, Published. towards Smartgrids". European Smart grids Technology
[7] Department of Energy, the Grid wise Alliance, the Galvin Platform: Vision and Strategy, Published, 13 Mar, 2006.
Electricity Initiative and EPRI/Intelligrid, ‘The Smart Grid: an
Introduction’ Published.
Abstract—This paper deal with the mission of solar Wind, solar, biomass, geothermal, tidal energy is
energy in India with there scope and challenges. Solar knows as renewable sources. Wind and solar is play a
energy is a renewable energy source which is used to major contribution to overcome the problem of
generate electrical energy by using solar PV. most of the electricity in India, but such sources are not used
country in the world generates their electricity by solar
energy. The demand of electricity increased day by day
effectively in India.
but the generation capacity of electricity is limited and this In India total number of solar plant in Gujarat is
can be overcome by using solar energy. A schematic plan maximum then other state in India.Gujrat is an example
is presented by this we can reduce the problem of in India for generation of electrical energy by using
electricity. solar energy effectively.
Keywords: Solar PV, climate condition, challenge and
scope.
I. INTRODUCTION
Indian power sector face a major problem, that is
the utilization of electrical energy is much more as
compare to generation and this can be over come if we
increase our dependency in renewable sources. Most of
the generation in India is fossil based then nuclear,
hydro and a small contribution by using renewable
sources.
Figg. 4: Solar PV
B. Climate Condition
Climatte conditionss play an im mportant role with
respect to solar generaation in Indiaa .whether chhange
time to tim
me which effeccts on the efficciency of the plant.
p
In India whhether is classsified into threee major part.
1. Winter.
W
2. Suummer.
3. Mansoon.
M
1) Winterr
The month
m of Noveember to Febbruary is know wn as
winter seasson. In this seeason the effeect of solar ennergy Fig. 7: Averagge Temperature iin India (Region Wise)
W
is reduced in comparisonn of summer, because sun is i not
C. Challenge-Laand
visible moost of the daays so the generation
g is also
effected .thhe average teemperature inn winter seasson is Is a scarce resource
r in Inndia and per capita land
about 100c-200c. availlability is low
w. Dedication oof land area fo
or exclusive
instaallation of solar arrays migght have to coompete with
2) Summer otherr necessities thhat require lannd.
The month
m of March to June iss know as sum mmer
D. Cost of Solar PV
season. Inn this season temperature increase andd this
period is best
b for generaation of electrricity. The average Cell is mucch higher, so initial inv vestment is
temperaturre in this seasoon is 350c-460c. increeased and Inddia is a develooping country
y so most of
the state
s is in deveeloping stage.
II. CONCLUSION
The problem of electricity reduced if we adopt non
conventional sources like solar energy.Goverment
support is necessary for awarance of solar PV in Indian
community. Other state also learns the utilization of
solar energy from Gujarat and, Rajasthan.
There is a need of commercialization of solar PV.
REFERENCES
[1] "Small Photovoltaic Arrays". Research Institute for Sustainable
Energy (RISE), Murdoch University. Retrieved 5 February
2010.
Fig. 8: View of Solar Park in Gujarat [2] John Quiggin (January 3, 2012). "The End of the Nuclear
Renaissance |". National Interest.
Next to Gujarat, Rajasthan is India's sunniest state, [3] People building their own solar systems from kits.
and many solar projects have been proposed. The 40 Greenplanet4energy.com. Retrieved on 2012-04-23.
[4] Example of diy PV system with pictures. Instructables.com
MW photovoltaic Dhirubhai Ambani Solar Park was (2007-11-05). Retrieved on 2012-04-23.
completed in April 2012.[18] A 250 MW compact [5] Deshgujarat.com/state wise-per-capita-power-consumption-
linear Fresnel reflector (CLFR) plant is under figures.
construction, consisting of two 125 MW. [6] Types of PV systems. University of Florida.
[7] List of Eligible SB1 Guidelines Compliant Photovoltaic
Modules.
keep the car ventilation going when parked in the sun. project on body panels that serve as a battery. The body
This adds to comfort and reduces the amount of energy panels are made of a sandwich material that is able to
needed to cool down the interior after return. store energy. The idea is to fight the 100kg penalty of
electric drive by giving the body this double function.
4) Biomaterials
Biomaterials have a long history in the automotive
industry. Leather upholstery has been there from the
start. Paneling has long been made of pressed wood
fibers and cotton and wool are well known as isolation
materials. New is the use of bioplastics. The technology
to produce bioplastics started with (non automotive)
degradable materials and has advanced to the point that
is possible to make parts up to automotive standards.
DSM has launched EcoPaXX in 2009 as a bio-based,
Fig. 2 high performance engineering plastic. Approximately
70% of the polymer consists of building blocks derived
2) New e-conductive materials
from castor oil. DSM proved EcoPaXX™ to be 100 %
Copper is heavy and increasingly expensive. The carbon neutral from cradle to gate. Castor beans are not
price of copper has seen a fourfold increase since 2003. fit for human consumption and hence the material does
The complete wiring harness of a car weighs not disturb the food production chain.
approximately 40 to 50 kg. The number of e- A relatively new trend in composites is to replace
connections tend to increase while multiplexing glass or carbon fibers with natural fibers from plants.
technology aims to bring the weight of the wiring There are various plants suited for this purpose. Most
harness down. New materials are sought for used is hemp. Hemp fibers can be cut to be used as
conductivity and insulation in wiring. Aluminum filler in thermoplast injection molded materials. The PP
instead of copper can save 4-7 kg on wiring per car but blend with 50% hemp fibers produced Greengran
has drawbacks on corrosion and needs a bigger cross creates parts that are stiffer, comparably strong, and
section (space). New insulation materials can reduce the cheaper then the conventional glass filled alternatives.
weight of a wire bundle by up to 40% (source: Delphi). The composite has fogging and humidity characteristics
Signal wires can be printed on foil to reduce weight and odor emission levels within present norms and a
and space. further advantage in better sound deadening. Hemp can
Conductivity of materials will also be a also be used as base for long fiber reinforced
development in the world of coatings. Transparent composites. The company Hempflax produced mats and
conductive layers are developed to create ice-free parts this way. As with Greengranfysical properties are
windshields. New conductive coatings are also the base very comparable to glass filled alternatives while these
of the OLED technology. products are half made from plants. The next step is the
resin. In 2010 DSM introduced a biobased resin for
3) Materials for energy storage automotive composites. Palapreg is 55% biobased and
New materials to store energy will be very this content of renewable material can be increased in
important for the future. The capacity and weight (and the years to come. Palapreg also offers a solution to the
costs) of batteries are still the major competitive recycling of composites. Since it contains short cycled
disadvantaged of electric vehicles compared to the carbon incineration can be regarded as CO2 neutral in
Internal Combustion Engine. The energy content of a terms of climate change. When combined with hemp
liter petrol is hard to beat. A fuel tank of 50kg will get a fibers this opens the perspective that composites
car 25 times as far as a Li-ion battery of 50kg and components can be fully recyclable.
refueling takes only a fraction of the time of recharging. OEMs that promote the use of biobased materials
Future technology will have to close this gap. Step by are Toyota, BMW and Daimler.
step. Many different ways are explored. The University
of Maryland reported (January 2011) that they are able III. CONCLUSION
to increase the capacity of a Li-ion battery by a tenfold The Netherlands have a strong raw material
by using a genetically modified virus that is attached to industry and a lot of design creativity. Hundreds of
and coated with special metals. The virus looses life Industrial Design graduated enters the market place
during the production process. each year. It is a chance for the automotive industry to
In the quest for better batteries also the attract them and give them a place to develop their
construction of the battery is bound to change. The skills. Their creativity can be put to use to generate new
Imperial Collage of London and Volvo have a research ideas for better products.
Design and Materials belong together. Many Dutch report that Dutch industry members can use to innovate
automotive companies work with materials, but many their product range are listed below:
more of them can benefit by using the creativity of • Monomaterial composites Styron.
designers. Design and materials shape the world around • Biobased materials DSM/ Greengran/
us. But designers can do much more. Designers are Hempflax.
trained to think from the customer point of view. They • PC glazing Sabic.
can think about how to improve the functionality of the • Fast reaction resins Styron.
product, also when it is a technical component. New • Solar/ oled/ smart foils Holst/ Solliance.
create thinking on functions can improve the product • New steel grades/ HSS Tata Steel/ M2i.
concept.
Thinking about the customer is thinking the REFERENCES
market. Designing products starts with thinking about
the future. Both in terms of market and technology. [1] 2100 Science–Shape Memory Alloys:
http://www.2100science.com/Videos/Shape_Memory_Alloys.as
Hence designers can support the product strategy px
process of the company. And when product [2] Cars on a Diet: The Material and Energy Impacts of Passenger
development gets technical lateral thinking of designers Vehicle Weight Reduction in the U.S / Lynette W. Cheah
can spark new ideas for new solutions. [3] Center for Transportation Analysis Research Brief “Automotive
Lightweight Materials Assessment”.
Design is a way to communicate quality. Design [4] Dutch Design Gallery; a gallery of Dutch automotive Designers
expresses care and dedication to the product. Also for [5] Innovation Norway: Advanced light weight materials within the
hidden components there is always a moment when this automotive industry light-Appendix 1 global trends–January
is important; when show in a brochure, at a sales 7th, 2010.
[6] Volvo-Tomorrow's Volvo car: body panels serve as the car
meeting when discussing price, at a show or even in the battery:
hands of the person who mounts it in the car to inspire http://www.volvocars.com/intl/top/about/corporate/volvo-
him to handle with care. sustainability/pages/sustainability-news.aspx?itemid=209
Design and materials are both a way to create a [7] Volkswagen-Stress-free driving through the winter:
http://www.volkswagenag.com/vwag/vwcorp/info_center/en/ne
better product. The Dutch automotive industry can ws/2010/11/volkswagen_group_researches_ice-
benefit from the results of strong Dutch material free_windscreens.html
suppliers, creative research of knowledge institutes and [8] Wikipedia-Shape memory polymer:
professional serve suppliers. A non exhaustive number http://en.wikipedia.org/wiki/Shape_memory_polymer
[9] WorldAutoSteel–Reinventing Automotive Steel (Greet Designs
of very promising developments mentioned in this in Steel–May5, 2010).
I. INTRODUCTION
Electricity generation using wind energy has been
well recognized as environmentally friendly, socially
beneficial, and economically competitive for many
applications. Because of crucial fossil energy resources
shortage and environmental issues the wind energy is
very important resource for electricity production.
A. WECS Technology Fig. 3: Main Elements of a Horizontal Axis Wind Turbine
A WECS is a structure that transforms the kinetic Figure 3 illustrates the major components
energy of the incoming air stream into electrical energy. placement in horizontal axis wind turbine.
This conversion takes place in two steps, as follows. A typical wind turbine consists of the following
The extraction device, named wind turbine rotor turns components:
under the wind stream action, thus harvesting a
mechanical power. The rotor drives a rotating electrical A. Blade
machine, the generator, which outputs electrical power.
Wind turbines are classified into two general types: An important part of a wind turbine that extracts
horizontal axis and vertical axis. A horizontal axis wind energy.
machine has its blades rotating on an axis parallel to the B. Hub
ground (Fig. 1). A vertical axis machine has its blades
rotating on an axis perpendicular to the ground (Fig 2). Blades are fixed to a hub which is a central solid
Today, the vast majority of manufactured wind turbines part of the turbine.
are horizontal axis with two or three blades, operating
C. Gear Box
either down-wind or up-wind.
Two types of gear box used in wind turbine:
1. Parallel shaft-It is used in small turbines,
design is simple, maintenance is easy, high
mass material and offset shaft.
2. Planetary shaft-It is used in large turbines,
complex design, low mass material and in line
arrangements.
D. Brakes
Two independent brakes sets or incorporated on the
rotor low speed shaft and high speed shaft.
The low speed shaft is Hydraulic operated .The
Fig. 1: Horizontal Axis Fig. 2: Vertical Axis wind high speed shaft is self adjusted and spring loaded.
wind Turbine Turbine
IV. POWER GENERATION SYSTEM Fig. 5: General Structure of a Limited Variable-Speed WECS
The electrical power generation structure contains
both electromagnetic and electrical subsystems. Besides VI. VARIABLE SPEED WECS
the electrical generator and power electronics converter Variable-speed wind turbines are currently the
it generally contains an electrical transformer to ensure most used WECS. The variable speed operation is
the grid voltage compatibility. possible due to the power electronic converters
interface, allowing a full (or partial) decoupling from
V. FIXED-SPEED WECS the grid.
Fixed-speed WECS operate at constant speed. That The doubly-fed-induction-generator (DFIG) based
means that, regardless of the wind speed, the wind WECS (Figure 6), also known as improved variable-
turbine rotor speed is fixed and determined by the grid speed WECS, is presently the most used by the wind
frequency. Fixed-speed WECS are typically equipped turbine industry.
The DFIG
D is a WRIG
W with thhe stator winndings
connected directly to the three phase, constant-
frequency grid and the rotor windinngs connectedd to a
back-to-baack (AC–AC) voltage sourcce converter. Thus, T
the term “doubly-fed”
“ comes from the fact thaat the
stator volttage is applieed from the grid g and the rotor
voltage is impressed by b the poweer converter. This
system alloows variable-sspeed operatioon over a largee, but
still restriccted, range, with the geenerator behaaviour
being goveerned by the powerp electroonics converteer and Fig. 7: General Structure
S of a Fulll Variable-Speed
d WECS
its controollers. The power elecctronics convverter
comprises of two IGBT T converters, namely the rotor VIII. CONTROL OPTIONS
side and thhe grid side coonverter, connnected with a direct
d
current (DC C) link. The rotor
r side connverter controls the
generator ini terms of acctive and reacctive power, while w
the grid sidde converter controls
c the DC-link
D voltage and
ensures opperation at a large power factor. The stator
outputs poower into thee grid all thee time. The rotor, r
depending on the operattion point, is feeding
f powerr into
the grid when
w the slip is negative (over
( synchroonous
operation) and it absorbbs power from m the grid wheen the
slip is possitive (sub-syynchronous operation). In both
cases, the power flow in the rotorr is approxim mately
proportionaal to the slip. DFIG-based WECS are highly h
controllablle, allowing maximum
m pow wer extractionn over
a large rangge of wind speeds.
Full variable-speed
v d WECS aree very flexibble in
terms of which
w type of generator is used.
u As presented
in Figure 4,
4 it can be eqquipped with eithere an induuction Fig. 8: Mechanical
M Controol & Load Contro
ol
(SCIG) or a synchronoous generator. The synchroonous
generator can be eitheer a wound-rrotor synchroonous A. Mechanical Control
C of the Turbine Blad
de
generator (WRSG)
( or a permanent-m magnet synchroonous
generator (PMSG).
( Thee back-to-backk power inverrter is As the wind speed changees the pitch off the blades
rated to thee generator poower and its operation
o is siimilar or bllade tip is adjjusted to conttrol the frequ
uency of the
to that in DFIG-based
D W
WECS. Its rotoor-side ensurees the turbiine rotation.
rotational speed being adjusted withhin a large range, r The drawback of this method iis that power in the wind
whereas its grid-side traansfers the acctive power to t the is wasted
w and coontrol methodd can be exp pensive and
grid and attempts too cancel thee reactive power p unreliable.
consumptioon. The PMSG G is a good option to be ussed in B. Load Controll
WECS, duue to its self-exxcitation propperty, which allows
a
operation at a high poweer factor and efficiency. PMSG As the wind speed changges the electriical load is
does not require
r energyy supply for excitation, ass it is channged by rapid switching, soo the turbine frequency
f is
supplied byb the permaanent magnetss. The stator of a contrrolled. This method
m makess greater use of o power in
PMSG is wound
w and the rotor has a permanent magnet
m the wind
w because the blade pitcch s kept at th
he optimum
pole systemm. The salientt pole of PMS SG operates at a low angle.
300 Bhag
gwant Institute of
o Technology, Muzaffarnagar,
M Uttar Pradesh, India
Wind Energy Conversion System
Abstract—The FACTS devices control the practical way to practice topologies of conversion of
interrelated parameters that rule the operation of the capable energy to give competitive solutions to the
transmission systems, including the serial impedance, the
related problems, with the interconnection of new
derivation impedance, the current, the tension, the phase
angle and the muffling of oscillations to different energy sources in the networks of high and medium
frequencies under nominal frequency. The FACTS tension and with improved the quality of the electricity
technology has a collection of controllers, that can be used supply.
individually or co-ordinated with other control installed in The concept of distributed generation is generally
the network, thus permitting to profit better of the associated to the development of the renewable energy
network’s characteristics of control. sources and other alternative sources as the fuel piles, is
The configurations of compensators based on another factor to keep in mind in the development and
switched inverters, called static compensators of reactive
configuration of the electrical system, that will need an
power, are today the most used in the electric current
transmission systems. The application of high powered important electronic equipment based on power
electronic converters makes possible the generation or converters that facilitate the integration of these sources
absorption of reactive power without using banks of of energy, without damaging over the reception quality
condensers or inductors. of the users connected to the electricity network.
The STATCOM is a device used on alternating The use of static power converters in electricity
current electricity transmission networks. It is based on a networks has the potential of increasing the capacity of
power electronic voltage-source converter and can act as transmission of electric lines and improving the supply
either a source or sink of reactive AC power to an quality of the electric energy. The devices used to
electricity network. If connected to a source of power it
achieve this, are the FACTS.
can also provide AC power. It is a member of the FACTS
family of devices. The function will depend on the type of According to the IEEE the definition of these
network in which it will be installed; it could fulfil FACTS devices is the following: ‘a power electronic
functions of reactive power control and power factor based system and other static equipment that provide
regulator, or other functions as shock absorber of system control of one or more AC transmission system
oscillations or filtering of harmonics, depending on the parameters to enhance controllability and increase
system applications. power transfer capability.’
Keywords: FACTS, STATCOM, Custom power, Active The FACTS controllers offer great opportunities to
filter. regulate the transmission of alternating current (AC),
increasing or diminishing the power flow in specific
I. INTRODUCTION lines and responding almost instantaneously to the
stability problems. The potential of this technology is
The solutions to improve the quality of supply in
based on the possibility of controlling the route of the
the electrical networks with distributed generation go
power flow and the ability of connecting networks that
through the application of the developments in are not adequately interconnected, giving the possibility
semiconductor power devices, that is to say, the of trading energy between distant agents.
utilization of static power converters in electrical
energy networks. The technological advances in power II. FACTS IN ACTIVE DISTRIBUTION SYSTEMS
semiconductors are permitting the development of
device that react more like an ideal switch, totally In the new market of Active Distribution, the
controllable, admitting high frequencies of flexibility of the transmission depending on the prices
commutation to major levels of tension and power. of the electric energy in each moment is imposed. The
On the other hand, large advances in auxiliary fact that the energy can vary dynamically in the way
technologies devices of digital control, DSP’s, circuits from generation to consumption requires a bigger
of programmable logic and in techniques of advanced margin of the lines design, or at least, a major control of
control. These technological developments, united to a the energy they transmit.
tendency of reduction of cost of the power Narrowly related to the FACTS term is that of
semiconductors, are permitting to undertake a new CUSTOM POWER, controllers based on solid state
technologies that are designed to operate in medium and Second generation: semiconductors with ignition
low tension levels, whose main objective is to improve and extinction controlled by door (GTO, IGBT, etc.).
the quality of service in distribution networks. The TABLE 1
solutions to improve the energy quality at the load side
FACTS Devices Attributes of Control
is of great important when the production processes get Static compensator of Voltage control and stability,
more complicated and require a bigger liability level, VAR’s (SVC, TCR, TCS, compensation of VAR’s, muffling of
which includes aims like to provide energy without TRS oscillations
interruption, without harmonic distortion and with Thyristor controlled series Currents control, muffling of
compensation oscillations, transitory, dynamic and of
tension regulation between very narrow margins. (TCSC, TSSC) tensions stabiles, limitations of fault
Among the FACTS, but that is different to them Current
because of their final use. In fact the topologies that Thyristor controlled Currents control, muffling of
they employ are identical to the ones in the FACTS Reactor series (TCSR, oscillations, transitory, dynamic and of
TSSR) tensions stabiles, limitations of fault
devices with little modifications and adaptations to Current
tension level; therefore they are most oriented to be Thyristor controlled Control of active power, muffling of
used in distribution networks of low and medium Phase shifting transformer oscillations, transitory, dynamics
tension, sometimes replacing active filters. (TCPST OR TCPR) transformer and of voltage stability
Thyristor controlled Control of reactive power, voltage
voltage regulator (TCVR) control muffling of oscillations,
III. ADVANTAGES AND OPERABILITY transitory, dynamics transformer and
OF FACTS DEVICES of voltage stability
Thyristor controlled Voltage limited limit of transitory and
The following features resume the main advantages voltage limited (TCVL) dynamics voltage
of the FACTS devices: Synchronous static Voltage control, compensation of
Better utilization of existing transmission system compensator (STATCOM VAR’s muffling of oscillations,
without storage) stability of voltage
assets: Cost of FACTS generally lower than that of new Synchronous static Voltage control and stability,
transmission lines. compensator (STATCOM compensation of VAR’s muffling of
Increased transmission system reliability and with storage) oscillations, transitory, dynamics and
of tension stability
availability: FACTS provide transmission systems with
Static synchronous series Current control, muffling of
robustness to endure contingencies. compensator (SSSC oscillations, transitory dynamics and of
Increased dynamic and transient grid stability: without storege) voltage stability, limitation of fault
Lower vulnerability to load changes, line faults. current
Increased quality of supply for sensitive industries: Static synchronous series Current control muffled of oscillations,
compensator (SSSC with transitory, dynamics and of tension
Through mitigation of flicker, frequency variations. storage) stability
Environmental protection: Smaller impact than the Unified power flow Control of active and reactive power,
installation of new lines. No waste production. controller (UPFC) voltage control, compensation of VAR
There are three factors to be considered before limitation of fault current
Interline power flow Control of reactive power, voltage
installing a FACTS device: controller (IPFC) or back control, muffling of oscillations
• The type of device. to back (BtB) transitory, dynamics and of voltage
• The capacity required. stability
• The location that optimize the functioning of First Generation Second Generation
304 Bhag
gwant Institute of
o Technology, Muzaffarnagar,
M Uttar Pradesh, India
FACTS Devicces in Distribute
ed Generation
Figg. 3: Statcom
A. SSSC (Static Synchrronous Series Compensatorr) The converteer 2 has thee main functtion of the
UPF FC; it injects an
a AC voltagee to the transmmission line,
This device
d works the same as the
t STATCO OM. It whicch magnitude and angle arre controllablee through a
has a volttage source converter
c seriially connected to seriaal transformer. The basic fuunction of the converter 1
a transmiission line through
t a trransformer. It is is to give or absoorb the real poower that the converter 2
necessary an energy soource to provvide a continnuous demaands in the common DC C link. The converter
c 2
voltage thrrough a conddenser and too compensatee the suppplies or absorbbs locally the required reacctive power
306 Bhag
gwant Institute of
o Technology, Muzaffarnagar,
M Uttar Pradesh, India
Procee
edings of Nation
nal Conference on Trends in Siignal Processing & Communica
ation (TSPC’13)), 12th–14th April 2013
M e on Clap onn
Move
Priti Verma1, Kannika Goyal2 and Richa Gupta
G 3
1,2
Deepartment of Electronics
E annd Communicaation,
Maharajaja Surajmal Institute
I of Tecchnology, Dellhi, India
3
Asst. Professsor, Departm
ment of Electro
onics and Commmunication,
Maharajaja Surajmal Institute
I of Tecchnology, Dellhi, India
richa1611@rediffmail.com
Abstract—The first ever World report on disab bility, II. HARDWARE SYSTEM
produced jointly by WH HO and the Woorld Bank, sugggests
that more than a billiion people in n the world today The hardwarre Design is based on ann embedded
experience disability. This paper describes the use u of systeem implemenntation usingg 8052 micrrocontroller
technology for speciallyy elderly and challenged peoplep fromm microchip. The
T block diaggram of hardw
ware system
almost everrywhere and ata anytime be it i be office, kittchen, is sh
hown in Fig 1.
bed room, study room through movee on clap on. This
paper dealss with motion sensors that track the moveement
of individu
ual along with h counter thatt count numb ber of
persons. In this wheneverr we enter in a room light an nd fan
d on. Now, fan
is switched n speed can bee regulated by clap.
The hardware and softwaare designed are oriented tow wards
microcontrroller based syystem, minimizzing the comp plexity
of system. The importan nt feature of this
t paper is tot use
clap sound to regulate thee speed of a fan
n.
Keywords: Microcoontroller, Bi-ddirectional M
Motion
Sensor and Clap Sensor Fig. 1: Block Diagram
D of Microocontroller Based
d Move
on Clap on D
Device
ISBN: 978-9
93-82880-20-2 307
Proceedings of National Conference on Trends in Signal Processing & Communication (TSPC’13) 12th–14th April 2013
Strictly speaking, individual PIR sensors do not The sound operated switch has the advantage that
detect motion; rather, they detect abrupt changes in the transmitter is always with us. Further it can be used
temperature at a given point. As an object, such as a in hospitals for handicapped and in houses too.
human, passes in front of the background, such as a These switches may be used to control lights or
wall, the temperature at that point will rise from room fans or anything else that does not produce too loud a
temperature to body temperature, and then back again. sound. To prevent an occasional loud sound from
This quick change triggers the detection. Moving causing malfunction, the circuit is normally quiescent.
objects of identical temperature will not trigger a The condenser microphone converts the sound of a
detection. clap into an electric signal and supplies it to the base of
PIDs can be equipped with more than one internal transistor BC148. The transistor provides a trigger pulse
sensing element so that, with the appropriate to monostable multivibrator IC NE555 timer. The
electronics, it can detect the apparent direction of duration for which the output of monostable
movement. As an object passes in front of adjacent multivibrator remains high is given by the relationship
sensors in turn, this implies the direction of movement. T = 1.1R*C*.
PIDs come in many configurations for a wide In this circuit, the values of R* and C* are so
variety of applications. The most common models have chosen that the clock pulse duration is approximately
numerous Fresnel lenses or mirror segments, an one second, so that no more than one clap within one
effective range of about ten metres (thirty feet), and a second can change the gear (and thus speed) of the fan.
field of view less than 180 degrees. Models with wider
fields of view, including 360 degrees, are available-- III. SOFTWARE SYSTEM
typically designed to mount on a ceiling. Some larger
PIDs are made with single segment mirrors and can This work is implemented using the following
software, Proteus–for designing circuit and simulation,
sense changes in infrared energy over one hundred feet
away from the PID. There are also PIDs designed with MPlab-for compilation, Embedded C–for programming
reversible orientation mirrors which allow either broad code, 8051 IDE–for dumping the programming code
into the microcontroller (Burner)
coverage (110° wide) or very narrow ‘curtain’
coverage, or with individually selectable segments to A. Microcontroller Software
‘shape’ the coverage.
In this case, the method consists of switching the
fan and lights by detecting motion and further
controlling the fan speed with sound of our clap. The
programming language used to program the
microcontroller is Embedded C. Many algorithms have
been investigated to choose the best fit method for the
microcontroller. The micro controller is programmed in
such a way that it takes input from the motion sensor
when anyone enters the room and displays the number
of people on seven segment display. Thus, switching
light and fan automatically by tracking the motion .
Now, fan speed can be regulated by clap.
Abstract—There has been exponential increase in which takes the input from the fingertip sensor and
health care cost in last decade. . Thus any improvements calculates the heart rate of the patient. The micro
in the diagnosis and treatment tools are welcomed by the controller also takes the responsibility todisplay the
medical community. This paper describes the same on LCD, which is interfaced to it through LCD
development of reliable, cheap and accurate heart beat
monitor system through heart beat sensor. This paper
drivers.SMS Module link will be used for sending the
deals with signal conditioning and data acquisition of captured diagnostic data[5].
heart rate signal. The hardware and software designed The next section giveshardware system overview;
are oriented towards microcontroller based system, SectionIII introduces software used for implementation
minimizing the complexity of system. The important of prototype fingertip heartbeat sensor. GSM Modem
feature of this paper is to use fingertip sensor to monitor explanation is given in IV. Result and Conclusion are
and compute heart rate to be displayed on LCD given in Section V, while future advancements are
monitor.It is also possible to interface the device with given in the end.
G.S.M. network to take assistance of a physician that is so
far from the place where the patient present and can
provide his advice for good treatment of the patient II. HARDWARE SYSTEM
according to his current condition of heart beat. The hardware Design is based on an embedded
Keywords: Heart beat Sensor, microcontroller, heart system implementation using PIC16F877A
beat monitoring, GSM Modem microcontroller from microchip. The block diagram of
hardware system is shown in Fig 1.
I. INTRODUCTION
Technology is being used everywhere in our daily
life to fulfill our requirements [1]. One of the ideal
ways of using technology is to employ it to sense
serious health problems so that efficient medical
services can be provided to the patient in correct time.
Changes in lifestyle and unhealthy lifestyles have
resulted in incidents of heart disease. Furthermore,
heart problems are being increasingly diagnosed on
younger patients. Hence there is a need that patient is
able to measure the hearth rate in home environment as
well.
Fig. 1: Block Diagram of Microcontroller based Heartbeat Monitor
The heart rate of a healthy adult at rest is around with Display on LCD
72 bpm. Athletes normally have lower heart rates than
less active people. Babies have a much higher heart rate The block diagram consists of Microcontroller
at around 120 bpm, while older children have heart PIC16F877A, Heart Beat Sensor, Reset,
rates at around 90 bpm. CrystalOscillator, LCDDriver, LCDDisplay, LCD
Heart beat monitor and display system is a portable Intensity Control, LED Indicators and a GSM Modem.
and a best replacement for the old model stethoscope
A. Microcontroller 16F877A
which is less efficient. The heart beat rate is calculated
manually using stethoscope where the probability of A Microchip Microcontroller PIC 16F877A [3] is
error is high because the heart beat rate lies in the range used to collect and process data. It has 256 bytes of
of 70 to 90 per minute whose occurrence is less than 1 EEPROM data memory, 2comparator, 8 channels 10
sec, so this device can be considered as a very good bit Analog to Digital Convertor. It has on chip 3 timers
alternative instead of a stethoscope. The functioning of and 8k flash program memory. The Heart beat Sensor
this device is based on the truth that the blood is interfaced to microcontroller via port pins. The
circulates for every heart beat which can be sensed by Output of Sensor is fed to microcontroller via ADC
using a circuit formed by the combination of an LDR (Analog to Digital Convertor).An LCD is used to
and LED. This device consists of a micro controller [2] display data.
This work
w is implemented usiing the following
software, Proteus–for
P deesigning circuuit and simulaation,
Fig. 4: The Idea of Transmission
T betw
ween Transmitter and Receiver
In ourr simple project, the progrram waits forr the dispplay the heart beat rate. Thhe goal of the paper is to
mobile nuumber to be entered throuugh the keyboard. reduuce the hospitaalization and assistance cosst.The pulse
When a ten t digit mobile number is provided,, the rate can be usedd to check ooverall heart health and
program innstructs the modem
m to sendd the text messsage fitneess level. Bessides it can pprove to be a boon for
using a seqquence of AT commands seniior citizen people
p who won’t have to travel
distaances or wait in long queuees at the hospiitals to get a
meaasure of their heart beat. They can themselves
hand dle this devicce easily by sitting at ho ome. Health
mon nitoring appliccation is mainnly proposed to provide
alertts for mediccal health m monitoring staaff for the
patieents when neeeded. The reall-time monitoring system
for cardiac
c patiennt physical staates is based on wireless
senssor network technology. It can be taken by
patieentand keep the
t patient moovement intacct because it
is miniature
m and portable.
p The system can monitor
m and
Fig. 5: Snappshot of the Deveeloped Module wiith GSM Technology recoord the physical states andd movement parameters
real--time, and theen provide ann auxiliary meeans for the
corrrect diagnosiis of doctorr. With the intelligent
diaggnosis softwarre, the sign off acute diseasee for patient
can be found earlly, and then tthe patient can n be helped
in tiime, the suddeen death of paatient can be avoided.The
a
low cost factor asssociated withh this device canc make it
a ho ousehold nam me. Temperatture sensor can c also be
inclu uded.
312 Bhag
gwant Institute of
o Technology, Muzaffarnagar,
M Uttar Pradesh, India
Proceedings of National Conference on Trends in Signal Processing & Communication (TSPC’13), 12th–14th April 2013
connected to the system; 4) reducing losses associated system without any type of compensation. The phasor
with the system; and 5) to prevent voltage collapse as diagram of these is also shown above. The active
well as voltage sag. The impedance of transmission current IP is in phase with the load voltage V2. Here,
lines and the need for lagging VAR by most machines the load is inductive and hence it requires reactive
in a generating system results in the consumption of power for its proper operation and this has to be
reactive power, thus affecting the stability limits of the supplied by the source, thus increasing the current from
system as well as transmission lines. Unnecessary the generator and through the power lines. Instead of
voltage drops lead to increased losses which needs to be the lines carrying this, if the reactive power can be
supplied by the source and in turn leading to outages in supplied near the load, the line current can be
the line due to increased stress on the system to carry minimized, reducing the power losses and improving
this imaginary power. Thus we can infer thatthe the voltage regulation at the load terminals. This can be
compensation of reactive power not only mitigates all done in three ways: 1) A voltage source. 2) A current
these effects but also helps in better transient response source. 3) A capacitor In this case, a current source
to faults and disturbances. In recent times there has device is used to compensate IQ, which is the reactive
been an increased focus on the techniques used for the component of the load current. In turn the voltage
compensation and with better devices included in the regulation of the system is improved and the reactive
technology, the compensation is made more effective. It current component from the source is reduced or almost
is very much required that the lines be relieved of the eliminated. This is in case of lagging compensation. For
obligation to carry the reactive power, which is better leading compensation, we require an inductor.Therefore
provided near the generators or the loads. Shunt we can see that, a current source or a voltage source can
compensation can be installed near the load, in a be used for both leading and lagging shunt
distribution substation or transmission substation. compensation, the main advantages being the reactive
power generated is independent of the voltage at the
II. COMPENSATION TECHNIQUES point of connection.
There are two types of compensation techniques.1) B. Series Compensation
Shunt compensation,2) Series compensation. The
principle of both Shunt and Series compensation are
described below:
A. Shunt Compensation
Fig. 3
3. We set a convergence criterion = i.e. if the inductance, the active power flows from the leading
largest of absolute of the residues exceeds, the source to the lagging source and the reactive power
process is repeated, or else its terminated. flows from the higher voltage magnitude source to the
4. Set the iteration count k = 0. lower voltage magnitude source. The phase angle
5. Set the bus count p = 1. difference between the sources determines the active
6. Calculate the real and reactive powers Pi and Qi…
power flow and the voltage magnitude difference
.Pi = Vi Vk Yik cos ( δi − δk − θik) between the sources determines the reactive power
flow. Thus, a STATCOM can be used to regulate the
Qi = Vi Vk Yik sin (δi − δk − θik)
reactive power flow by changing the magnitude of the
7. Compute the schedule errors ΔPI and ΔQI. VSC voltage with respect to source bus voltage.
ΔPir =Pisp– Pi r(cal)
ΔQir = Qi sp– Qir (cal)
If the calculated value of Qi is within limits, only
ΔPiis calculated. If the calculated value of Qi is beyond
the limits, then an appropriate limit is imposed and
ΔQiis also calculated by subtracting the calculated value
of Qi from the appropriate limit
8. Compute the elements of the jacobian matrix
Fig. 4
using the estimated |Vi| and δifrom step 2.
9. Obtain Δδ and Δ|Vi|.
V. PROPOSED WORK
10. Using the values of Δδi andΔ|Vi|, modify the
voltage magnitude and phase angle at all load In this paper the per unit voltage and angle of 30
buses. bus system is used to carry out the comparative study.
11. Start the next iteration cycle at step 2 with these This comparative study is done using Newton Raphson
modified |Vi| and δi. algorithm with and without STATCOM.The per unit
12. Continue until scheduled errors ΔPir and ΔQir for bus voltage and angle plot with and without STATCOM
all load buses are within a specified tolerance.
is shown below:
IV. STATIC COMPENSATOR (STATCOM) A. Without STATCOM
One of the many devices under the FACTS family, xnlf=[1 1.0600 0.0000
a STATCOM is a regulating device which can be used 2 1.0430 -5.3543
3 1.0196 -7.5308
to regulate the flow of reactive power in the system 4 1.0104 -9.2840
independent of other system parameters. STATCOM 5 1.0100 -14.1738
has no long term energy support on the dc side and it 6 1.0096 -11.0581
cannot exchange real power with the ac system. In the 7 1.0020 -12.8649
8 1.0100 -11.8193
transmission systems, STATCOMs primarily handle 9 1.0392 -14.0644
only fundamental reactive power exchange and provide 10 1.0215 -15.6706
voltage support to buses by modulating bus voltages 11 1.0820 -14.0644
during dynamic disturbances in order to provide better 12 1.0496 -15.1245
13 1.0710 -15.1245
transient characteristics, improve the transient stability 14 1.0320 -16.0018
margins and to damp out the system oscillations due to 15 1.0251 -16.0084
these disturbances. A STATCOM consists of a three 16 1.0304 -15.6251
phase inverter (generally a PWM inverter) using SCRs, 17 1.0188 -15.8687
18 1.0114 -16.6067
MOSFETs or IGBTs, a D.C capacitor which provides 19 1.0066 -16.7658
the D.C voltage for the inverter, a link reactor which 20 1.0095 -16.5502
links the inverter output to the a.c supply side, filter 21 1.0082 -16.2178
components to filter out the high frequency components 22 1.0120 -15.9811
23 1.0085 -16.2294
due to the PWM inverter. From the d.c. side capacitor, a 24 0.9991 -16.3007
three phase voltage is generated by the inverter. This is 25 1.0032 -16.0720
synchronized with the a.c. supply. The link inductor 26 0.9852 -16.5038
links thisvoltage to the a.c. supply side. This is the basic 27 1.0145 -15.6559
28 1.0078 -11.7163
principle of operation of STATCOM 29 0.9944 -16.9077
For two AC sources which have the same 30 0.9828 -17.8067]
frequency and are connected through a series
Fig. 5
Fig. 6
Abstract—This paper studies about the research on Despite the power of the search strategies, it is very
ways to extend and improve query methods for image difficult to optimize the retrieval quality of CBIR within
databases is widespread. we have developed the QBIC only one query process. The hidden problem is that the
(Query by Image Content) system to explore content- extracted visual features are too diverse to capture the
based retrieval methods. To achieve the high efficiency
and effectiveness of CBIR we are using two type of
concept of the user’s query. To solve such problems, in
methods for feature extraction like SVM(support vector the QBE system, the users can pick up some preferred
machine)and NPRF(navigation-pattern based relevance images to refine the image explorations iteratively. The
feedback).By using svm classifier as a category predictor feedback procedure, called Relevance Feedback (RF),
of query and database images, they are exploited at first to repeats until the user is satisfied with the retrieval
filter out irrelevant images by its different low-level, results. The rest of the paper is organized as follows: A
concept and key point-based features. Thus we may review of past studies is briefly described in Section 2.
reduce the size of query search in the db then we may In Section 3, we describe the details of our proposed
apply NPRF algorithm and refinement strategies for method. Empirical evaluations of the proposed method
further extraction then we may combine the color and the
texture of the given query image to Obtain optimal
are expressed in Section 4. Finally, we conclude the
solution. paper in Section 5.
Keywords: navigation-patterns, color based approach, II. PRIOR AND RELATED WORK
texture based approach, SVM
Relevance feedback [5], [2], [12], in principle,
I. INTRODUCTION refers to a set of approaches learning from an
assortment of users’ browsing behaviours on image
With the advent of imaging, care could be
retrieval [10]. Some earlier studies for RF make use of
significantly impacted with improved image handling.
existing machine learning techniques to achieve
In recent years, rapid advances of software and
semantic image retrieval, including Statistics, EM,
hardware technology have eased the problem of
KNN, etc. Although these forerunners were devoted to
maintaining large medical image collections. These
formulating the special semantic features for image
images constitute an important source of anatomical
retrieval, e.g., Photo book [11], QBIC [1], Visual SEEK
and functional information for the diagnosis of diseases,
[16], there still have not been perfect descriptions for
medical research, and education. Effectively and
semantic features. This is because of the diversity of
efficiently searching in these large image collections
visual features, which widely exists in real applications
poses signif-general histogram equalization (GHE) and
of image retrieval. Therefore, active query refinement,
local histogram equalization (LHE) [1]–[4], [5]
based on the analysis of usage logs, attracts researchers’
brightness preserving dynamic histogram equalization
attention in this area of RF.
(BPDHE) or SVE.BPDHE is obtained from dynamic
It is very difficult to derive an adaptive and perfect
histogram specification [10] which generates the
measuring function. A specific measuring function
specified histogram dynamically from the input image.
indeed cannot cover all target groups with various
The proposed method can remove impulse noise of any
visual contents. Moreover, the modified query point of
density from the satellite images and enhance the
each feedback probably moves toward the local optimal
contrast of the image.
centroid. Thus, global optimal results are not easily
Content-Based Image Retrieval (CBIR) is the
touched in QPMlike work spreading in the broad
mainstay of current image retrieval systems. In general,
the purpose of CBIR is to present an image feature space. As a result, diverse results for the same
conceptually, with a set of low-level visual features concept are difficult to obtain. For this reason, the
such as color, texture, and shape. These conventional modified version of MARS [9] groups the similar
approaches for image retrieval are based on the relevant points into several clusters, and selects good
computation of the similarity between the user’s query representative points from these clusters to construct the
and images via a query by example (QBE) system [11]. multipoint query.
318 Bhag
gwant Institute of
o Technology, Muzaffarnagar,
M Uttar Pradesh, India
Study of Retrieval Method based on Query by Image Content
Precision =
Recall =
Even a single relevant image should be missed in
the output as well as to minimize the number of
irrelevant images.
Fig. 5
REFERENCES [8] H.T. Shen, S. Jiang, K.L. Tan, Z. Huang, and X. Zhou, “Speed
up Interactive Image Retrieval,” VLDB J., vol. 18, no. 1, pp.
[1] A. Pentalnd, R.W. Picard, and S. Sclaroff, “Photobook: Content- 329-343, Jan. 2009.
Based Manipulation of Image Databases,” Int’l J. Computer [9] V.S. Tseng, J.H. Su, J.H. Huang, and C.J. Chen, “Integrated
Vision (IJCV), vol. 18, no. 3, pp. 233-254, June 1996. Mining of Visual Features, Speech Features and Frequent
[2] T. Qin, X.D. Zhang, T.Y. Liu, D.S. Wang, W.Y. Ma, and H.J. Patterns for Semantic Video Annotation,” IEEE Trans.
Zhang, “An Active Feedback Framework for Image Retrieval,” Multimedia, vol. 10, no. 2, pp. 260-267, Feb. 2008.
Pattern Recognition Letters, vol. 29, pp. 637-646, Apr. 2008. [10] V.S. Tseng, J.H. Su, B.W. Wang, and Y.M. Lin, “Web Image
[3] J.J. Rocchio, “Relevance Feedback in Information Retrieval,” Annotation by Fusing Visual Features and Textual Information,”
The SMART Retrieval System—Experiments in Automatic Proc. 22nd ACM Symp. Applied Computing, Mar. 2007.
Document Processing, pp. 313-323, Prentice Hall, 1971. [11] K. Vu, K.A. Hua, and N. Jiang, “Improving Image Retrieval
[4] Y. Rui, T. Huang, and S. Mehrotra, “Content-Based Image Effectiveness in Query-by-Example Environment,” Proc. 2003
Retrieval with Relevance Feedback in MARS,” Proc. IEEE Int’l ACM Symp. Applied Computing, pp. 774-781, 2003.
Conf. Image Processing, pp. 815-818, Oct. 1997. [12] L. Wu, C. Faloutsos, K. Sycara, and T.R. Payne, “FALCON:
[5] Y. Rui, T. Huang, M. Ortega, and S. Mehrotra, “Relevance Feedback Adaptive Loop for Content-Based Retrieval,” Proc.
Feedback: A Power Tool for Interactive Content-Based Image 26th Int’l Conf. Very Large Data Bases (VLDB), pp. 297-306,
Retrieval,” IEEE Trans. Circuits and Systems for Video 2000.
Technology, vol. 8, no. 5, pp. 644-655, Sept. 1998. [13] P.Y. Yin, B. Bhanu, K.C. Chang, and A. Dong, “Integrating
[6] J.R. Smith and S.F. Chang, “VisualSEEK: A Fully Automated Relevance Feedback Techniques for Image Retrieval Using
Content-Based Image Query System,” Proc. ACM Multimedia Reinforcement Learning,” IEEE Trans. Pattern Analysis and
Conf., Nov. 1996. Machine Intelligence, vol. 27, no. 10, pp. 1536-1551, Oct. 2005.
[7] G. Salton and C. Buckley, “Improving Retrieval Performance by
Relevance Feedback,” J. Am. Soc. Information Science, vol. 41,
no. 4, pp. 288-297, 1990.
Abstract—In this paper the authors have proposed a efficiency of the device decreases due to tunneling
simple method to study the change in the millimeter-wave induced phase distortion. A generalized method of
performance of IMPATTs operating in MITATT mode small-signal analysis based on Gummel-Blue technique
from the shift of avalanche transit time (ATT) phase [5] was first reported in [6], where the authors carried
delay. A generalized double iterative computer method
based on Gummel-Blue approach incorporating the effect
out simulation on the high frequency negative resistivity
of tunneling is used to obtain admittance characteristics profiles in the depletion layer of the IMPATT devices.
and negative resistivity profiles of silicon based double The effect of tunneling was incorporated in the above
drift region (DDR) device structures operating at different reported small-signal analysis by Dash and Pati [7] to
mm-wave frequencies in the presence of tunneling. It is study the high frequency admittance characteristics and
observed that a shift of ATT phase delay occurs when negative resistivity profiles of IMPATT devices
tunneling is taken into account. This shift is obtained from operating in MITATT mode.
the spatial variation of the negative resistivity profiles in In this paper the authors have proposed a simple
the depletion layer of the device. The results show that the method of calculating the tunneling induced shift of
shift of ATT phase delay increases at higher operating
frequencies of the device, which in turn leads to larger
ATT phase delay from a study of shift of negative
degradation of the mm-wave performance of the device at resistivity peaks due to tunneling in the depletion layer
higher frequencies. of the device. This method of obtaining shift of ATT
phase delay in DDR IMPATTs due to the effect of
Keywords: Admittance characteristics; ATT phase tunneling is described in the next section. Using this
delay; MITATT; Millimeter-wave; Tunneling.
method the tunneling induced ATT phase delays of
silicon DDRs operating at different mm-wave
I. INTRODUCTION GHG frequencies are calculated and results are presented in
The tunneling current in IMPATT device operating section III. Finally the paper is concluded in section IV.
at higher millimeter-wave frequencies causes
considerable degradation of RF performance of the II. METHOD
device. The thinner depletion layer (W < 0.80 μm) of
The avalanche transit time (ATT) phase delays of
mm-wave IMPATT device and the higher peak electric
the mm-wave DDR devices can be calculated from the
field (Em > 6×107 V/m) near the junction are the corresponding R(x) profiles for the following two cases
favorable conditions for tunnel generation of electron from which the shift of ATT phase delay due to
hole pair (EHP) provided vacant states are available in tunneling current can be computed: (i) without
the conduction band opposite to the filled states of the considering the tunneling current; which gives ATT
valance band. Both the tunnel generated and the phase delay in pure avalanche mode and (ii) considering
avalanche generated carriers move in the drift region to the tunneling current; which gives ATT phase delay in
produce the necessary transit time delay for IMPATT mixed mode.
device is known as mixed tunneling avalanche transit The R(x) profiles of the devices for the above two
time mode or MITATT mode. W. T. Read was first cases exhibit negative specific resistance peaks in the
considered the tunnel current in IMPATT operation in drift layers, but the magnitudes and locations of these
his famous paper [1]. Other researchers [2] followed by maxima change when tunneling effect is taken into
him carried out the theoretical analysis of IMPATT account in MITATT mode.
device based on the simplifying assumption of equal The spatial shift of the negative specific resistance
ionization rates of electrons and holes in the respective maxima for a particular base material determines the
semiconductors. Elta and Haddad [3] used an effective shift of ATT phase delay of MITATT due to tunneling
ionization rate of charge carriers in their analysis on current. If the distances of the peaks from the junction
MITATT device by introducing the concept of dead on the p-side of ATT device in the avalanche and mixed
space. Luy et al [4] considered a time dependent tunnel mode are xpa and xpm and the corresponding optimum
current and observed that the DC to RF conversion frequencies fa and fm respectively, then the phase delays
on the p-side of the device at xpa and xpm for cases (i) frequency of the device increases in MITATT mode of
and (ii), are obtained from the following relations: operation.
Figure 1 shows the peak avalanche generation rate
(1) and peak tunneling generation rate versus frequency (in
MITATT mode) graphs. It can be observed from Figure
Similarly the phase delays on the n-side of the 1 that initially at the lower frequencies avalanche
device can be obtained from the following relations: generation dominates over tunneling generation. But at
higher frequencies (above 380 GHz) tunneling
generation begins to dominate over avalanche
(2)
generation. Another thing can be observed from Figure
1 and Table II that the ratio of the peak tunneling
generation rate to the peak avalanche generation rate
(gTpeak/gApeak) increases with operating frequency of
the device. It is interesting to note that the degradation
of the RF performance of the device is strongly related
to the gTpeak/gApeak – ratio of the corresponding device.
As the frequency of operation of the device in MITATT
mode increases, gTpeak/gApeak – ratio also increases which
The shifts of ATT phase delays (in MITATT leads to more degradation of the mm-wave performance
device) due to the effect of tunneling are determined of the device.
from the following relations:
(3)
(4)
The results presented in the next section show that
the effect of tunneling causes an effective shift of the
R(x) profile towards the p-side of the device. That is
why on the p-side of the device, ϕ pm > ϕ pa ; which
means δp is positive. But on the n-side of the device
ϕnm < ϕna; which means δ n may be negative or
positive. Generally p-side of the device is
predominantly affected by the tunneling than the n-side
of that [7]. Thus the value of δ p is greater than the
value of δ n (i.e. δp / δn > 1 ). The overall shift of ATT
phase delay in MITATT device is obtained by
averaging over δp and δn given by,
Fig. 1: Peak Avalanche Generation Rate and Peak Tunneling
Generation Rate Versus Frequency
(3)
Figure 2 shows the small-signal admittance
Double Drift Region (DDR) IMPATT devices characteristics or conductance-susceptance plots of the
based on silicon are designed and optimized for CW devices in both IMPATT and MITATT modes. It can be
operation at different millimeter-wave frequencies. The observed from Figure 2 that the magnitude of negative
method of analysis presented in [6-7] is used to conductance decreases in the MITATT mode, which in
simulate the DC and RF properties of DDR IMPATT turn reduces the RF power output of the device. Figure
devices designed for optimum performance. The design 2 also shows that the optimum frequency of the device
parameters of the devices are given in Table I. shifted upward in MITATT mode of operation.
Table III shows the simulated DC and small-signal These decrement of magnitude of negative
parameters of the devices in both IMPATT and conductance and upward shift of operating frequency in
MITATT modes. It can be observed that all the DC MITATT mode is more prominent at higher
parameters like peak electric field, breakdown voltage, frequencies. Table III shows that the percentage of
DC to RF conversion efficiency as well as all the small decrement of magnitude of negative conductance and
signal parameters like magnitude of peak negative percentage of increment of optimum frequency are
conductance, negative resistance, RF power output are 6.4% and 24.5% respectively in the device operating at
decreased in MITATT mode of operation of the 122GHz, while the same are 8% and 60% respectively
devices. Another interesting fact is that the optimum in the device operating at 480 GHz.
III. CONCLUSION
A Simple method is proposed by the authors in this
paper to calculate the tunneling induced shift of ATT
phase delay in MITATT device. Results show that as
the frequency of operation of the device increases the
shift of ATT phase delay due to tunneling increases
which leads to greater degradation of the RF
performance of the device.
REFERENCES
Fig. 3: Negative Resistivity Profiles of the Devices in both IMPATT
and MITATT Modes [1] W.T. Read, “A proposed high-frequency negative resistance
diode”, Bell Syst. Tech. J., vol. 37, pp. 401-466, 1958.
Figure 3 shows the negative resistivity profiles of [2] M. Chive, E. Constant, M. Lefebvre and J. Pribetich, “Effect of
the devices in both IMPATT and MITATT modes. tunneling on high efficiency IMPATT avalanche diode”, Proc.
IEEE (Lett.), vol. 63, pp. 824-826, 1975.
Magnitude of negative resistivity at each space point [3] M.E. Elta and G.I. Haddad, “Mixed tunneling and avalanche
in the depletion layer of the device decreases in mechanism in p-n junctions and their effects on microwave
MITATT mode and negative resistivity profiles are transit time devices”, IEEE Trans. Electron Devices, vol. 25, pp.
slightly shifted towards the p-side of the device. This 694-702, 1978.
[4] J.F. Luy and R. Kuehnf, “Tunneling assisted Impatt operation”,
spatial shift of the negative resistivity profile in IEEE Trans. Electron Devices, vol. 36, pp. 589-595, 1989.
MITATT mode is actually the measure of the tunneling [5] H.K. Gummel and J.L. Blue, “A small-signal theory of
induced shift of ATT phase delay. The authors have avalanche noise in IMPATT diodes”, IEEE Trans. on Electron
evolved a simple method for calculating the tunneling Devices, vol. 14, no. 9, pp. 569-580, 1967.
[6] S.K. Roy, J.P. Banerjee and S.P. Pati, “A computer analysis of
induced shift of ATT phase delay from a study of the the distribution of high frequency negative resistance in the
shift of negative resistivity peaks due to tunneling in the depletion layers of impatt diodes” Proc. of NASECODE-IV
depletion layer of the device. The principle of Conf. on Numerical Analysis of Semiconductor Devices
calculating shift of ATT phase delay due to tunneling (Dublin: Boole Press), pp. 494, 1985.
[7] G.N. Dash and S. P. Pati, “A generalized simulation method for
is outlined in section II. The magnitudes of ATT MITATT-mode operation and studies on the influence of tunnel
phase delay shifts for silicon based MITATT devices current on IMPATT properties”, Semicond. Sci. Technology,
operating at different mm-wave frequencies are vol. 7, pp. 222-230, 1992.
calculated and plotted against the optimum frequency
of the corressponding devices in Figure 4. It can be
TRACK VI
ABSTRACTS
ISBN: 978-93-82880-20-2 3
Proceedings of National Conference on Trends in Signal Processing & Communication (TSPC’13), 12th–14th April 2013
In paper investigation that Philosophy for Children of pupils in classroom dialog technique , and improved
stems from the work analysis by Neetu Singh for children reasoning in justification of opinions. There
primary school students. It involves teacher modeling of were no changes in comparison classes. Increases were
exploratory talk in relation to a complex stimulus then found in the amount of teacher-student dialog and the
has been students discuss together in pairs, larger amount of student-student dialog, with decreases in the
groups, and the whole class to achieve consensus amount of teacher-whole-class talk. This study found
through advanced technology systems. The effects of gains in cognition which were sustained from
Philosophy for Children on quantity and quality of elementary school into high school even when no
interactive more type dialog on 5- 10-year old children further Philosophy for Children was done. Socio-
in primary mainstream classes in (Agra )India. the emotional aspects of learning has been also improved.
investigated Video recordings of teacher-led and pupil- Investigation of process found changes in teacher
pupil classroom dialog before, and six months into, behavior and child behavior. Implications for future
participation in the program were analyzed. Changes in research, policy and practice are outlined, particularly
intervention classes included increased use of open- for enhancing consistency and expanding the program.
ended questions by the teacher, increased participation
The success or failure of the entire software improving the efficiency of the process. This paper
development process relies on software testing activity analyses the effectiveness of applying genetic algorithm
which is responsible for ensuring that the software is (GA) for generating test data automatically using data
free from bugs. One of the major labor intensive flow testing approach. The GA conducts its search by
activities of software testing is the automatic generation constructing new test data from previously generated
of test data that satisfy a given adequacy criterion for test data that are evaluated as effective test data.
the purpose of applying the testing methodologies. Finally, the paper presents the results of the
Many approaches have been tried and tested for experiments that have been carried out to evaluate the
automating the process of generating the test data. effectiveness of the proposed GA.
Meta-heuristics have been applied extensively for