Project Catalogue
Project Catalogue
Project Catalogue
Undergraduate/Graduate Projects
Catelogue
Contents
II Hardware Design
1
. . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . .
8
8
IV Biometrics
10
V Data Mining
10
12
Data Processor
Event Correlator
. . . . . . . . . . . . . . . . . . . . . . . . . . .
14
14
. . . . . . . . . . . . . . . . . . . .
12
. . . . . . . . . . . . . . . . . . . . . . . . . . . .
13
VI Radio Communication
1
12
16
.
16
VII Cryptography
16
17
. . . . . . .
18
. . . . . .
. . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . .
18
19
20
20
20
the calculations, the storage to store the data and the network between a combination of these two.
(bottleneck) for the overall performance. Current trend shows that the processor's power is not used to full capacity because the data to be processed is not
retrieved fast enough from the hard disks. Hence, to make read requests faster,
a storage system that caches the data with an ecient (caching) algorithm
(strategy) is required.
In this project, members are expected to design and develop an in-memory
(i.e in-RAM) database engine that is optimized for a specic database whose
table schemas and application behaviour are going tobe provided by INSA.
PROJECT_TYPE: Graduate
(a) Building a search engine may entail some acquaintance with data
mining techniques like TF-IDF
PROJECT_TYPE: Graduate | Undergraduate
References
1. BIG-DATA Platform & BIG-DATA Analytics Research & Development
Experimental Development Approaches in HPC and Data Mining
2. Ghemawat, Sanjay, Howard Gobio, and Shun-Tak Leung. "The Google
le system." ACM SIGOPS operating systems review. Vol. 37. No. 5.
ACM, 2003.
3. Hindman, Benjamin, et al. "Mesos: A Platform for Fine-Grained Resource
Sharing in the Data Center." NSDI. Vol. 11. 2011.
and easiest way of programming Xilinx FPGAs directly from CPU or microcontrollers without using JTAG cables. In this project, it is expected to design
a system which consists of a microcontroller (for parallel slave mode programming), and two ash storages (one for program, and another one for data storage).
PROJECT_TYPE: Undergraduate
4 Design and implementation of software dened radio (SDR) using FPGA processors 7
technology is predicted to replace many of the traditional methods of implementing transmitters and receivers while oering a wide range of advantages
including adaptability, re-congurability and multi-functionality encompassing
modes of operation, radio frequency bands, air interfaces, and waveforms. The
most computationally intensive part of a SDR is the channelizer, which extracts
multiple narrowband channels from a wideband input signal. The objective of
this project is to implement and design the channelizer using FPGA processor.
PROJECT_TYPE: Graduate | Undergraduate
Linux based host machines only, and it is not integrated into any of IDEs.
However,most developers usually prefer to develop applications using IDEs like
eclipse. Because IDEs provide easy build and debugging interfaces and, fast and
exible code editors. It also allows program portability across dierent host architectures and operating systems as well. The objective here is to develop SDK
plugin for Eclipse IDE.
PROJECT_TYPE: Undergraduate
ables users to manage home appliances (light, security cameras ,TV, cooking
appliances,refregirator and many others) remotely using mobile phones. Basic
requirements of the system are:
10
Some n-
Wavelet analysis
and its applications to ngerprint verication is one of the fast growing areas
for research in recent year. Wavelet based technique results in high recognition
rates. The use of wavelet in image feature extraction lies in the fact that the
wavelet transform clearly separates the high frequency and low frequency information on a pixel by pixel basis.
and frequency domain and often give a better signal representation using Multiresolution analysis. Wavelet transform based method takes less response time
which is more suitable for online verication with high accuracy, reliable and
better technique. The ngerprint recognition methods in wavelet transform domain converts the ngerprint images directly to the feature vector form, and
then verify it. It preserves a potential of developing a lost cost, and small size
ngerprint recognition system.
In this project, a student/group of students will undertake a literature review of the various published studies in the area of wavelet based ngerprint
recognition, analyze and select the most promising algorithms, implement the
algorithms using C/C++ and nally evaluate their implementation using online free test databases. The project does not need additional materials except
a personal computer.
PROJECT_TYPE: Graduate | Undergraduate
11
anti-spoof problems should be well solved before face recognition could be widely
applied in our life.
performance are based on intensity images and equipped with a generic camera.
Thus, an anti-spoong method without additional device will be preferable,
since it could be easily integrated into the existing face recognition systems.
The aim of the project will be studying, selection, implementation and testing
of state-of-the-art liveness detection methods in face recognition. Only a web
cam will be enough to successfully nish the project. Alternatively, online free
test databases can be exploited for a more scientic and hassle free research
project.
PROJECT_TYPE: Graduate | Undergraduate
References
1. Rafael C.Gonzalez and Richard E. Woods, "Wavelets and Multi-resolution
processing," Digital Image Processing, Third Edition.
2. A. Graps., An introduction to wavelets, IEEE Comput.
Sci.
Engr.,
2(2):50-61, 1995.
3. Online: Robi Polikar, The engineer's ultimate guide to wavelet analysis,
4. U. Uludag and A. K. Jain. Attacks on biometric systems: A case study
in ngerprints.
In Proc.
12
pivotal role in preserving the intellectual property of the country and making
it easily accessible for the posterity.
The
data can be - structured data (Eg - XML), unstructured data (Eg - Websites, Blogs and forums . . . ) or semi structured data.
3 Data Processor
13
compact datasets that retains only the most important features of the
original document.
4. Sentiment analysis: Responsible for detecting the contextual polarity of
text; that is, it determines whether a piece of writing is positive, negative
or neutral.
PROJECT_TYPE: Graduate | Undergraduate
3 Data Processor
Data preprocessor is a tool with the objectives of integrating data from multiple
sources into a coherent data format, transforming the integrated data into a form
that is suitable for analysis and reducing this data into a smaller representation,
yet closely maintaining the integrity, of the original data. This tool has methods
and algorithms under each category (integration, transformation, reduction) in
order to address the dierent kinds of problems and requirements that mass
data has and the analysis requires respectively:
1. Data Cleaning - Attempts to ll in missing values, smooth out noise while
identifying outliers, and correct inconsistencies in the data. Under Data
Cleaning we have: Missing Values, Noisy Data and Cleaning.
2. Data Integration - Attempts to combine data from multiple sources into
a coherent data format (store) Under Data Integration we have: schema
integration, object matching and redundancy control.
3. Data Transformation - Attempts to transform or consolidate the data into
forms that are appropriate for analysis. Under Data Transformation we
have:
construction.
4. Data Reduction Data - Attempts to obtain reduced representation of the
data set that is much smaller in volume, yet closely maintains integrity
of the original data.
i.e.
more ecient yet produce the same (or almost the same) analytical result.
Under Data Reduction we have: attribute subset selection, dimensionality reduction, numerosity reduction, discretization and concept heirarchy
generation.
PROJECT_TYPE: Graduate | Undergraduate
4 Event Correlator
14
4 Event Correlator
Event correlation is a procedure where a stream of events from multiple sources,
such as applications like anti viruses, rewalls, intrusion detection systems,
servers and so on, is processed with a range of correlation methods in order
to detect (and act on) certain event groups that occur within predened time
windows. The event correlator has dierent techniques and operations to correlate events, in real time, coming either from only same sources or multiple ones,
individual events or groups of events, similar events or dierent kinds of events
and so on:
1. Correlation Techniques - Under this category we have - rule based event
correlation, nite state machine based, case based reasoning, model based
reasoning, bayesian network based event correlation, neural network approaches.
2. Correlation Operations - Under this category we have - compression, logical operation, aggregation, ltering, suppression, thresholding, rate limiting, escalation, temporal relations, generalization, specialization, clustering.
PROJECT_TYPE: Graduate | Undergraduate
References:
1. Christopher D. Manning, Prabhakar Raghavan and Hinrich Schtze, Introduction to Information Retrieval, Cambridge University Press. 2008.
2. Diapanjan Das, Andre F.T. Martins, A survey on automatic Text Summarization, Nov 21, 2007.
3. Evans D.K Similarity Based Multi-lingual multi-document summarization
Columbia University 2005.
15
4. Feng Hao John Daugman, Piotr Zielinsky, A fast search algorithm for
alarge fuzy database, IEE Transaction on information forensic and security, june 2008
5. H. Bast, A,Chitea, F.M. Shutanek and I Weber.Ester : Eceint search on
text,entitites and relations. In SIGIR, Pages 671-678, 2007.
6. Data Mining, Concepts and Techniques (Jiawei Han and Micheline Kamber), Chapter Two
7. Event Correlation Engine (Master's Thesis), Andreas Muller, Spring Term
2009
8. Weka Data Mining Software (Open Source)
16
At the
receiver end, the eect of the channel should be estimated and compensated
(equalized) so as to regain the transmitted signal. The received signal will be
demodulated and detected into series of symbols which then will be decoded to
retrieve the original message information.
The goal of this project is to select reliable and ecient algorithms for equalization, channel estimation, demodulation and detection to fulll the functional
receiver which retrieve original message from digitally modulated signal.
For
this, the modulation types are limited to - Phase modulation (QPSK, 8PSK,
16PSK), Frequency modulation (FSK, GMSK), Amplitude and phase modulations (ASK2PSK8). The decoding module is applicable to known coding techniques such as PCM coded audio, or ASCII 7/8 formatted text.
17
phones and tablets, which have smaller network loads but also much smaller
central processing units (CPUs) and limited battery life.
Sometimes, rather
than not deploying cryptographic protection at all, programmers react to performance problems by deploying low-security cryptography. The reason is that
many cryptographic libraries allow trade-os between security and performance.
Any government needs a trusted cryptographic base from which it can (legallyforce to) build all security solutions for its communication and storage systems.
The Reasons are 1. To avoid or close weak links that usually happen as a consequence of a
naive choice of cryptographic algorithms or cryptographic systems by ordinary application and system developers with no background in cryptography. Those naive choices are; ofcourse, a consequence dierent inherent
tradeos of system attributes including but not limited to - security, simplicity (of implementation and usage), performance (speed) and energy
(power consumption in large systems).
2. To Protect a data of diverse nature which can be systematically categorized into the following two classses
(a) Resident_Data - Any data that is processed by exactly one program. Example - Document processors - word/microsoft-word2007,
PDF/adobe- pdf, etc.
(b) Transmitted_Data - Any data that is processed by two or more
programes.
18
constrained RFID systems with respect to memory, power, and energy of the
tags are main challenges to provide security and privacy in RFID system's applications. Cryptographic techniques are commonly used to enable authentication
while protecting privacy and security. Unfortunately, these algorithms and their
corresponding implementations are dicult to adapt to the extreme conditions
implied by the use of RFID. The extremely limited budget for energy and area
do not allow the use of traditional cryptography. Many lightweight protocols
have been proposed so far that are founded on both, private- and public-key
cryptosystems.
19
Reference
1. Concept Paper: INSA Crypto-Platform, Concept Paper, INSA, Information Assurance Directorate
2. Securing Communication:
20
source code
2. ACTION/LOGIC
load the right parser plugin for the input source code
load the right vulnerabily(s) database for the right parser plugin
21
3. OUTPUT