SPECTRUM SENSING

TECHNIQUES AND
APPLICATIONS
SPECTRUM SENSING
TECHNIQUES AND
APPLICATIONS

MARCELO SAMPAIO DE ALENCAR,

FABRCIO BRAGA SOARES DE CARVALHO,
AND
WASLON TERLLIZZIE ARAJO LOPES

MOMENTUM PRESS, LLC, NEW YORK

Spectrum Sensing Techniques and Applications

Copyright Momentum Press, LLC, 2018.

All rights reserved. No part of this publication may be reproduced, stored

in a retrieval system, or transmitted in any form or by any means
electronic, mechanical, photocopy, recording, or any otherexcept for
brief quotations, not to exceed 400 words, without the prior permission
of the publisher.

First published by Momentum Press, LLC

222 East 46th Street, New York, NY 10017
www.momentumpress.net

ISBN-13: 978-1-60650-979-1 (print)

ISBN-13: 978-1-60650-980-7 (e-book)

Momentum Press Communications and Signal Processing Collection

Collection ISSN: 2377-4223 (print)

Collection ISSN: 2377-4231 (electronic)

Cover and interior design by Exeter Premedia Services Private Ltd.,

Chennai, India

10 9 8 7 6 5 4 3 2 1

Printed in the United States of America

This book is dedicated to our families.
 Abstract

Spectrum Sensing deals with several subjects that range from S tatistical
and Probability Theory to Radio Propagation and Signal Processing,
and cognitive radio is important to the evolution and dissemination of
new applications on spectrum sensing. The objective of this book is to
present a connection among basic statistical formulation, the fundamental
concepts from signal detection and spectrum sensing, and its integration
to cognitive radio and dynamic spectrum access, filling in the gaps from
previous books and leading to an interesting, robust, and illustrative

content, with recent practical applications of cognitive radio and spectrum
sensing. Recent applications based on spectrum sensing are presented
including some fundamental distribution probabilities for the mathemat-
ical presentation of spectrum sensing theory, and examples, illustrations,
and graphics help the reader understand the theory.

KEYWORDS

cognitive radio, cognitive networks, software-defined radio, spectrum

sensing, wireless sensor networks
 Contents

List of Figures xi
List of Table xiii
List of Symbols xv
List of Acronyms xvii
Preface xxi
Acknowledgments xxiii
1Cognitive Radio 1
1.1Introduction 1
1.2 Fundamentals of Cognition 2
1.3 History of Cognitive Radio 3
1.4 Software-Defined Radio Fundamentals 4
1.5 Definition of Spectral Holes 4
1.6 Cognitive Networks Architecture 6
1.7 Operation of Cognitive Networks 9
1.8 Regulation and Standards 11
1.9 Spectrum Efficiency 16
1.10 Underlay, Overlay, and Interweave 17
1.11 Cognitive Radio Definitions 19
2Spectrum Sensing and Detection 23
2.1Introduction 23
2.2Fundamentals of Spectrum Sensing and Detection 24
2.3 Hypotheses Tests 25
2.4 Signal Detection and Estimation 28
2.5 Spectrum Sensing Methods 30
2.6 Energy Detection Spectrum Sensing 31
2.7Spectrum Sensing Based on Statistical Tests 36
xContents

3Distributed Sensing 41
3.1Introduction 41
3.2 Cooperative Spectrum Sensing 43
3.3 Distributed Spectrum Sensing 50
4Spectrum Sensing Applications 53
4.1 TV White Spaces 53
4.2 Vehicular Networks 54
4.3Smart Grid and Power Line Communications 62
4.4Public Emergency and Health Care Networks 64
4.5 Fifth Generation Cellular Systems 64
4.6 Wireless Sensor Networks 65
5Important Distributions Used in Spectrum Sensing Theory 69
5.1 Exponential Distribution 69
5.2 Laplace Distribution 69
5.3 Gaussian Distribution 71
5.4 The Chi-square Distribution 72
5.5 Gamma Distribution 73
5.6 Rayleigh Distribution 74
5.7 Rice Distribution 75
5.8 Nakagami Distribution 77
5.9 Lognormal Distribution 77
5.10 The von Mises Distribution 78
5.11 Direction of Arrival Distributions 80
5.12 Pareto Distribution 81
5.13 Weibull Distribution 82
About the Authors 85
References 89
Index 101
 List of Figures

Figure 1.1.The dynamics of the spectrum hole concept.  3

Figure 1.2.Representation of the spectral holes. 5
Figure 1.3.Cognitive radio conceptual architecture. 7
Figure 1.4.The system model for femtocell-based
cognitive radio n etworks. 11
Figure 1.5. Spectrum overlay and underlay. 18
Figure 1.6.Block diagram of a software defined radio. 20
Figure 2.1.Primary and cognitive users operating in the same
geographical area. 24
Figure 2.2.Block diagram for the spectrum sensing based on
energydetection. 33
Figure 3.1.The hidden terminal problem in cognitive networks. 42
Figure 3.2.Cooperative spectrum sensing (with fusion center). 43
Figure 3.3. Distributed spectrum sensing (no fusion center). 50
Figure 3.4.Parallel topology for distributed spectrum sensing. 51
Figure 3.5.Sequential topology for distributed spectrum sensing. 52
Figure 4.1.Example of vehicular communications. 55
Figure 4.2.V2V and V2I communications. 57
Figure 4.3. Opportunistic communication in a vehicular network. 59
Figure 4.4. Noncentric CRV architecture. 60
Figure 4.5. Limited infrastructure CRV architecture. 60
Figure 4.6. Complete infrastructure support architecture. 61
Figure 5.1. Exponential probability density function, with
parameter a = 1. 70
Figure 5.2. Laplace probability density function, with parameters
a = 1 and b = 0. 70
xii List of Figures

Figure 5.3.The Gaussian probability density function, for three

different values of the mean, , and standard variation,
X.72
Figure 5.4.Chi-squared probability density function. 73
Figure 5.5. Gamma probability density function. 74
Figure 5.6. Rayleigh probability density function. 75
Figure 5.7. Lognormal probability density function. 78
Figure 5.8. The von Mises distribution for several values of . 79
Figure 5.9.Comparison between the pdfs of von Mises, Gaussian,
and cosine for Q = 4, = 27, = 4.7. 81
Figure 5.10.Envelope probability density function for = 1.0, for
different values of the parameters and . Circles
represent Monte Carlo simulation, for integer. 84
 List of Table

Table 1.1.Comparison of various indoor wireless access

technologies10
 List of Symbols

R0Acceptance Region
ADAnderson-Darling
hFading Amplitude
WSignal Bandwidth
RBayesian Cost
R1Critical or Rejection Region
Decision Threshold
tDecision Time
PdDetection Probability
Pd,cDetection Probability for Cooperative Sensing
YEDEnergy Detection Decision Estimation
E[]Expected Value Operator
u0Final Decision
Error of Type I
Error of Type II
()Eulers Gamma Function
PfaFalse Alarm or False Detection Probability
Pfa,cFalse Alarm or False Detection Probability for Cooperative Sensing
H0Hypothesis that the Channel is Idle
H1Hypothesis that the Channel is Occupied
JBJarqueBera
KSKolmogorovSmirnov
KKurtosis
mXStatistical Mean
PmdMissed Detection Probability
Pmd,cMissed Detection Probability for Cooperative Sensing
I0Modified Bessel Function of Zero Order
uNoncentrality Parameter
NNumber of Samples
TObservation Interval
xvi List of Symbols

2Pearson Chi-square
SWShapiro-Wilk
Signal-to-Noise Ratio
SSkewness
BkSpectrum Resource from Primary Network k
(y) Test Statistics
2Variance
 List of Acronyms

AP Access Point
ADC Analog-to-Digital Converter
ADSL Asymmetric Digital Subscriber Line
AM Amplitude Modulation
AMI Advanced Metering Infrastructure
Anatel Brazilian National Telecommunications Agency
AR Adaptive Radio
AWGN Additive White Gaussian Noise
BPSK Binary Phase Shift Keying
BS Base Station
CBTS Cognitive Base Transceiver Station
CDR Constant Detection Rate
CEM Coverage Extension of Macro
CEPT European Conference of Postal and Telecommunica-
tions Administrations
CFAR Constant False Alarm Rate
COA Closed or Open Access
COH Closed or Open or Hybrid
CR Cognitive Radio
CRSN Cognitive Radio Sensor Networks
CRV Cognitive Radio Vehicular ad hoc networks
CR-VANET Cognitive Radio Vehicular ad hoc NETworks
CR-WSN Cognitive Radio-Wireless Sensor Networks
CSMA Carrier Sense Multiple Access
DAC Digital-to-Analog Converter
DCO DSL or Cable or Optic fiber
DDC Digital Down-Conversion
DL Downlink
DoA Direction of Arrival
DSA Dynamic Spectrum Allocation
DSL Digital Subscriber Line
xviiiList of Acronyms

DSP Digital Signal Processor

DSRC Dedicated Short-Range Communications
DSSS Driving Safety Support Systems
DUC Digital Up-Conversion
DySPAN Dynamic Spectrum Access Networks
ED Energy Detector or Energy Detection
ECC Electronic Communications Committee
ECDF Empirical Cumulative Distribution Function
FBS Femtocell Base Station
FC Fusion Center
FCC Federal Communications Commission
5G Fifth Generation
FM Frequency Modulation
FSU Femtocell Secondary User
FFT Fast Fourier Transform
GHOST Goodness of Fit Higher Order Statistics Testing
GLRT Generalized Likelihood-Ratio Test
GoF Goodness of Fit
HAN Home Area Networks
HSO Hot Spot or Office
HSOT Hot Spot or Office or Tunnel or High-Speed Train
IEEE Institute of Electrical and Electronic Engineers
IF Intermediate Frequency
IHO In-Home or Office
IHS In-Home or Hot Spot
IoT Internet of Things
IR Intelligent Radio
ISM Industrial, Scientific, and Medical
ISO International Organization for Standardization
ITS Intelligent Transportation Systems
ITU-R Radiocommunication Sector of the International
Telecommunication Union
ITU-T Telecommunication Standardization Sector
LAN Local Area Network
LRT Likelihood-Ratio Test
LTE-A Long-Term Evolution - Advanced
MAC Medium Access Control Layer
MANET Mobile Ad Hoc Network
MBAN Medical Body Area Network
ML Maximum-Likelihood
MSU Macrocell Secondary User
 List of Acronyms xix

NP NeymanPearson
OA Open Access
OBU On-Board Unit
OFCOM Federal Office of Communications
OFDM Orthogonal Frequency Division Multiplexing
OI Outdoor and Indoor
OSI Open Systems Interconnection
PBS Primary Base Station
PDAs Personal Digital Assistants
pdf Probability Density Function
PHY Physical Layer
PLC Power Line Communications
PU Primary User
QoS Quality of Service
QPSK Quadrature Phase Shift Keying
RF Radio Frequency
ROC Receiver Operating Characteristic
RSU Roadside Unit
RTVD Real-Time Voice and Data
RV Random Variable
SCR Software-Controlled Radio
SDR Software-Defined Radio
SNR Signal-to-Noise Ratio
SU Secondary User
3GPP Third-Generation Partnership Project
TV Television
UHF Ultra High Frequency
UL Uplink
U-NII Unlicensed National Information Infrastructure
UWB Ultra-Wideband
VANET Vehicular Ad Hoc Network
VHF Very High Frequency
VoIP Voice over Internet Protocol
V2I Vehicle-to-Infrastructure
V2P Vehicle-to-Person
V2R Vehicle-to-Roadside
V2V Vehicle-to-Vehicle
WAN Wide Area Networks
WAVE Wireless Access in Vehicle Environment
WIF Wireless Innovation Forum
Wi-Fi Wireless Fidelity
xxList of Acronyms

WIO Wireless In-band or Out-of-band

WLAN Wireless Local Area Network
WRAN Wireless Regional Area Network
WSN Wireless Sensor Network
X2 X2 Interface
 Preface

Spectrum Sensing is a new topic in Communications Engineering, which

deals with different subjects, that range from Statistical and Probability
Theory to Radio Propagation and Signal Processing. Owing to the con-
tinuous advance in wireless communications, computer engineering, and
signal processing, a novel and important subject is dynamic spectrum
sensing. In this scenario, cognitive radio is important to the evolution and
dissemination of new applications on spectrum sensing.
The employment market in this area is in demand of information tech-
nology professionals and engineers who could afford to change and learn,
as the market changes. The market needs professionals who are able to
model and design complex systems.
Cognitive spectrum sensing is a recent topic on Wireless Communi-
cations, and few books have been published covering the subjects needed
to understand the fundamental concepts of dynamic spectrum sensing and
cognitive radio applications. Most books that deal with the subject are
destined to specific topics or strategies.
The more mathematically oriented books are destined to people with
engineering and statistical background, because the authors are more
interested in theorems and related conditions, than in fundamental con-
cepts and applications. The books written for engineers usually lack the
required mathematical rigor, or skip some important points in favor of
simplicity and conciseness.
The objective of this book is to present a connection among basic
statistical formulation, the fundamental concepts from signal detection
and spectrum sensing, and its integration to cognitive radio and dynamic
spectrum access, filling in the gaps from previous books and leading to an
interesting, robust, and illustrative content, with recent practical applica-
tions of cognitive radio and spectrum sensing.
The book begins with the recent evolution of cognitive radio and the
evolving of dynamic spectrum access. It also introduces the basic c oncepts
xxiiPreface

of software-defined radio, detection theory, spectrum sensing, and the

main methods described in the literature, which are basic to a course on
spectrum sensing. Additionally, advanced spectrum sensing strategies,
such as cooperative and distributed sensing, are detailed.
Recent applications based on spectrum sensing are presented as
well as some fundamental distribution probabilities for the m athematical
presentation of spectrum sensing theory. Examples, illustrations, and

graphics help the reader understand the theory.

The first chapter introduces the concepts of cognitive radio and pres-
ents software-defined radio fundamentals, along with the definitions of
spectrum holes. The cognitive network architecture is presented and the
operation of cognitive networks is discussed, including topics on regula-
tion and standards.
The fundamentals of spectrum sensing and detection is the topic of
Chapter 2, which also discusses the ideas of licensed and cognitive users,
presents some tests of hypothesis, related to signal detection and estima-
tion, and explains the main spectrum sensing methods.
Chapter 3 deals with cooperative and distributed sensing, discusses
some fusion strategies, distributed spectrum sensing, and distributed sens-
ing topologies.
Spectrum sensing applications are presented in Chapter 4, along with
the topics of cognitive vehicular networks, wireless sensor networks, and
its use in precision agriculture and industrial networks.
The most important probability distributions used in spectrum sens-
ing theory are discussed in Chapter 5.
Marcelo Alencar, Fabrcio Carvalho, and Waslon Lopes
 Acknowledgments

The authors are grateful to all the members of the Communications

Research Group, certified by the National Council for Scientific and Tech-
nological Development (CNPq), at the Federal University of Campina
Grande and at the Federal University of Paraiba, for their collaboration in
many ways, helpful discussions and friendship, as well as our colleagues
at the Institute for Advanced Studies in Communications (Iecom).
The authors are thankful to professor Orlando Baiocchi, from the
University of Washington, Tacoma, USA, who strongly supported this
project from the beginning and helped with the reviewing process.
 CHAPTER 1

Cognitive Radio

1.1INTRODUCTION

The allocation of the electromagnetic spectrum among several telecom-

munications services is a complex task. Different technologies, based
on the transmission of guided waves or free space propagation share the
available frequency bands. The Radiocommunication Sector of the Inter-
national Telecommunication Union (ITU-R) is responsible for recommen-
dations regarding the international frequency allocation.
Several countries also established regulatory agencies, for telecom-
munication services and distribution of frequency bands. These agencies
manage the spectrum occupation for different services. The Federal Com-
munications Commission (FCC) in the United States, the National Telecom-
munications Agency (AnatelAgncia Nacional de Telecomunicaes) in
Brazil, and the Federal Office of Communications (OFCOM) in Switzer-
land are examples of these agencies (Alencar and da Rocha Jr. 2005).
In general, international regulatory authorities frequently allocate
only one service per frequency band. As new services are licensed, new
frequency bands are also licensed (Tandra et al. 2009). This allocation
policy is static and inadequate for the current scenario, in which new tech-
nologies are competing to occupy the spectrum.
A survey of the recent literature indicates that the electromagnetic
spectrum is crowded in some spectrum bands, while other bands are vacant
or present scarce use (Raychaudhuri and Mandayam 2012). In the United
States the spectrum occupation ranges from 15% to 85% in some fre-
quency channels (Umar and Sheikh 2013). Measurements of the spectral
use, during a three year period in the city of Chicago, shown an average
occupation of only 14% in the band from 30 MHz to 3 GHz. The FM radio
frequency (between 87.9 MHz and 107.9 MHz) presented an a verage occu-
pation of about 80%, while the Industrial, Scientific, and Medical (ISM)
band at 2.45 GHz presented an average occupation of only 20%.
2 SPECTRUM SENSING TECHNIQUES AND APPLICATIONS

The consolidation of digital television systems around the world

pushed some countries to turn off the analog transmission of TV signals.
Countries, such as the United States, in 2009, and part of the European
Union, in 2012, disallocated frequency bands in the Very High Frequency
(VHF) and Ultra High Frequency (UHF) ranges (Carvalho et al. 2015c).
Because of the nonefficient spectrum allocation in combination of the
availability of frequency bands from the analog TV, the FCC proposed to
regulate the Dynamic Spectrum Allocation (DSA), to allow an opportunis-
tic access to the spectrum (Carvalho et al. 2015c).
The users in the system are classified as primary and secondary, or
cognitive users (Ghasemi and Sousa 2007). The Primary Users (PU) are
those who acquired a license to use a specific frequency band, allocated by
the national regulatory agency, while the Secondary Users (SU) can oppor-
tunistically use the frequency band (Akyildiz et al. 2006a).
The cognitive users should be able to monitor the frequency spectrum
and, from the observations, determine if there is a primary user occupying
a certain frequency band. The verification of the spectral occupancy is
possible, thanks to the cognitive radio emergence.

1.2FUNDAMENTALS OF COGNITION

The cognitive radio technology promises some new and interesting advan-
tages for radio users. In this regard, Software-Defined Radio (SDR) is a
natural platform on which to build in new cognitive features (SDR Forum
2008), and some new concepts emerge from the amalgamation of commu-
nications and software.
An Adaptive Radio (AR) is a communications system that monitors
its own operation, using a predefined set of metrics and rules, and modifies
certain operating parameters to improve its performance. It is the basis
upon which cognitive and intelligent radios are designed.
Cognitive Radio (CR) is a technique for wireless communication, in
which a transceiver can intelligently, another anthropomorphism widely
used in the area, detect the communication channels that are in use, and
transmit using blank, or vacant, channels, also called spectrum holes, while
avoiding occupied ones. This optimizes the use of the available radio fre-
quency (RF) spectrum and minimizes the interference with other receivers.
Figure 1.1 illustrates the concept of spectrum holes (Akyildiz et al. 2006b).
Intelligent Radio (IR) is a type of cognitive radio that is capable of
machine learning. This allows the cognitive radio to improve the ways in
which it adapts to changes in performance and environment to improve
the quality of service of the end user.
 Cognitive Radio3

Radio environment

Transmitted RF stimuli
signal
RF stimuli

Spectrum
decision Spectrum sensing
Information on
spectrum holes

Channel Information on
capacity spectrum holes

Spectrum
analysis

Figure 1.1. The dynamics of the spectrum hole concept.

Based on (Akyildiz et al. 2006b).

1.3 HISTORY OF COGNITIVE RADIO

The current concept of cognitive radio was proposed by Joseph Mitola III,
in a seminar at the Royal Institute of Technology in Stockholm, in 1998,
and published in an article by Joseph Mitola III and Gerald Q. Maguire,
Jr., in 1999 (Mitola III and Gerald Q. Maguire 1999).
Cognitive radio was considered a novel approach in wireless commu-
nications, which Mitola later described as:

The point in which wireless Personal Digital Assistants (PDAs)

and the related networks are sufficiently computationally intelli-
gent about radio resources and related computer-to-computer com-
munications to detect user communications needs as a function of
use context, and to provide radio resources and wireless services
most appropriate to those needs.

It is usually seen as an objective toward which an SDR platform should

evolve: a reconfigurable wireless transceiver that automatically adapts its
communication parameters to support different network functions and
user demands.
4 SPECTRUM SENSING TECHNIQUES AND APPLICATIONS

1.4 SOFTWARE-DEFINED RADIO FUNDAMENTALS

A cognitive radio is defined as a transmitting equipment that is aware of

the surrounding radioelectric environment. Based on software-defined
objectives, it can implement decisions that lead to use the available spec-
trum in the most appropriate way.
Cognitive radio is a hybrid technology that involves concepts from
SDR and Wireless Sensor Network (WSN). Possible functions of cogni-
tive radio include the ability of a transceiver to determine its geographic
location, identify and authorize its user, encrypt or decrypt signals, sense
neighboring wireless devices in operation, adjust output power, and
change modulation characteristics.
There are two main types of cognitive radio, full cognitive radio and
spectrum sensing cognitive radio. Full cognitive radio takes into account
all parameters that a wireless node or network can be aware of. Spectrum
sensing cognitive radio is used to detect channels in the RF spectrum. It can
be done as a two-layer mechanism, Physical (PHY) layer sensing, which
focuses on detecting the primary user signal, and Medium Access Control
(MAC) layer sensing, which determines the channels that should be sensed
by the secondary users, to minimize the sensing delay (Xiang et al. 2010).
Cognitive radio uses a number of technologies that includes AR, in
which the signal processing system monitors and modifies the perfor-
mance of the communications equipment, and SDR, in which the usual
hardware devices, including filters, mixers, modulators, and amplifiers are
replaced by computer routines.
In the context of WSNs, the cognitive radio is a tool that allows to
change the transmission parameters from the interaction of the radio and
the environment (Akyildiz et al. 2008). The use of cognitive communica-
tions contributes with the reliability of the network as well as with the cost
of the sensors (Vijay et al. 2010). The cognitive networks could provide,
to the users, a large bandwidth by using techniques to dynamic access of
the electromagnetic spectrum (Sousa 2013).

1.5 DEFINITION OF SPECTRAL HOLES

In general, the cognitive radio may be expected to deal with parameters

such as channel occupancy, available channels, the type of data to be trans-
mitted, and the modulation types that may be used. It must also consider
the requirements and rules established by the regulatory bodies.
The cognitive radio is able to identify spectral holes, defined as the
frequency bands that are allocated to a primary user but, for a given time
 Cognitive Radio5

and a given location, are not used by those users, or attend a minimum
interference criterion (Haykin 2005).
The spectral holes can occur in time, in frequency, or space ( Tandra
et al. 2009). Models for spectral occupancy take into consideration those
factors (Lopez-Benitez and Casadevall 2014). Information about the prop-
agation of the primary signal and about the localization of the cognitive
user allows the extension of the sensing to other dimensions (Wei et al.
2015, Wijenayake et al. 2013). Requirements, regarding the mobility of
cognitive radios or the polarization of the transmitted waves, are also
investigated to permit a suitable spectral detection (Guo et al. 2013, Paura
and Savoia 2013).
The classification of the RF bands, for cognitive use, can be per-
formed according to the power spectrum of the radio signals as follows
(Haykin 2005):

White spacesfrequency bands without interference, except the

one caused by environmental noise.
Gray spacesfrequency bands that can be occupied with low
power interference.
Black spacesfrequency bands occupied by transmissions with
high power interference.

Figure 1.2 illustrates the occupation of the spectral holes. The cog-
nitive radio technology allocates, dynamically and opportunistically,

Power Frequency Spectrum in use

Dynamic access to
the spectrum

Time

Spectrum holes

Figure 1.2. Representation of the spectral holes. Based on

(Akyildiz et al. 2009).
6 SPECTRUM SENSING TECHNIQUES AND APPLICATIONS

d ifferent network users in licensed and nonlicensed frequency bands. This

allocation could occur in frequency, time, or in power. When a certain
band of the electromagnetic spectrum is idle in a specific time, that band
could be handed off to a secondary user. The licensed user has priority for
the use of the frequency channel previously allocated and, if the primary
user needs to use that band, the cognitive user should immediately leave
that frequency band.

1.6 COGNITIVE NETWORKS ARCHITECTURE

From the definition, two main characteristics of the cognitive radio could
be highlighted: its cognitive capacity, to identify parts of the electromag-
netic spectrum that are not in use in a given time or place, and its recon-
figuration capacity (Rocha 2013). Thus, the cognitive radio technology
allows its users (Sousa 2013):

To determine the available frequency bands of the electromagnetic

spectrum, and to detect the presence of primary users in a frequency
band (spectral sensing).
To select the best available channel to perform a transmission
(spectral management).
To allow the channel access to other users (spectral sharing).
To keep the communication of the secondary user, when a primary
user is detected, while the communication is transferred to other
band (spectral mobility or spectral handoff).

Therefore, it is possible to identify two major subsystems in a cogni-

tive radio. A cognitive unit, that makes decisions based on specific inputs,
and a flexible SDR unit, whose software provides a range of possible oper-
ating modes. A separate spectrum sensing subsystem, which supervises
the transmissions, is also often included in the architecture, to measure the
signal environment and determine the presence of other services or users
(Akyildiz et al. 2008)
It is important to note that those subsystems do not necessarily define
a single device, but may include components that are distributed through-
out the entire network. Therefore, a cognitive radio is often referred to as
a cognitive radio system, or as a cognitive network.
The cognitive unit can be separated into two main parts, as shown in
the block diagram of Figure 1.3 (SDR Forum 2008). One part, labeled the
cognitive engine, searches a solution or optimizes a performance objective
 Cognitive Radio7

User Domain

Application
CE
Internal Monitoring O Transport
and System Policy S
External Inputs Rule-Set I
S Network
T
Cognitive Engine
A Link/MAC
Control System Policy Engine C
Operational
K
Rule-Set PHY

Radio Domain

Policy Domain

Figure 1.3. Cognitive radio conceptual architecture. Based on (SDR Forum

2008).

based on internal or external inputs received, which define the radios cur-
rent internal state and operating environment (Akan et al. 2009).
The Open Systems Interconnection (OSI) protocol stack houses a
family of information exchange standards developed jointly by the Inter-
national Organization for Standardization (ISO), and the Telecommunica-
tion Standardization Sector (ITU-T).
Another important part is the policy engine, which guarantees that the
solution provided by the cognitive engine complies with regulatory rules
and other policies that govern the network. The police engine is connected
to the cognitive engine, the radio domain subsystem, and the OSI stack.
The provision of support for cognitive radio and dynamic spectrum
access requires certain enabling technologies, which are under develop-
ment by the members of the Wireless Innovation Forum (WIF), such as
(Xiang et al. 2010):

Information Process ArchitectureOne of the key questions is how

to understand the current state of complex information systems, and
their associated communications subsystems, to determine how to
improve their procedures, from a process perspective, and how to
analyze them to increase the chances to interact with other systems.
An information process architecture provides a top-down model,
and a series of tools, to represent the structure of complex systems,
8 SPECTRUM SENSING TECHNIQUES AND APPLICATIONS

to help define, design, and select relevant cognitive radio processes,

to improve the understanding of the structure, and the relationships
between different systems that cross user domains.
Modeling LanguageEfficient and adjustable communication
protocols are required, between advanced radio systems and sub-
systems, to support next generation features, such as vertical and
horizontal mobility, spectrum awareness, dynamic spectrum adap-
tation, waveform optimization, advanced modulation schemes, fea-
ture exchanges, and other advanced applications.
A modeling language, which is built on specific aspects of use
cases, and that defines the signaling plan, the requirements and the
technical analysis of the information exchanges, can be the basis to
develop specifications and standards to support the required capa-
bilities.
Radio Environment MapThe operation of a cognitive engine
requires data and metadata to define the spectral environment in
which terminal operates at a given instant of time. The radio envi-
ronment map can include information on spectrum economic trans-
actions, dropouts, handovers, available networks, and services.
The information contained in the map is partially obtained
through the synthesis of measurements captured from many radios.
It may be stored in a database that can be accessed remotely by the
cognitive engine to aid the decision process.
Requirements for a database structure that implements this access,
including standardized database structures, data formats, and func-
tionality must be created, to support the flexibility necessary to accom-
modate current and future cognitive radio spectrum applications.
Test and MeasurementCognitive radios present particular test
challenges to quantify the performance of critical functions such as
spectrum sensing, interference avoidance, database performance,
and adherence to defined policies.
Test methodologies that support these challenges must consider a
set of metrics, a range of hardware platforms, protocols, algorithms,
use cases, and spectrum stakeholder requirements. Test equipment
functionality and performance, test interfaces, and test modes must
also be taken into account.

Other requirements, in addition to the level of processing required for

cognitive radio, include flexible RF sections, because they may need to
swap frequency bands, possibly moving between parts of the RF s pectrum
 Cognitive Radio9

that are widely separated, and they also may need to change between
transmission modes that could occupy different bandwidths.
To achieve the required level of performance it is necessary to have
a very flexible front-end. Traditional front-end technology cannot handle
those requirements because they are generally band limited. Also, the
modulation scheme used is generally fixed as well as the frequency band
in which it operates.
Regular wideband receivers present limitations, and generally o perate
by switching front-ends, when required. Accordingly, the required level
of performance can only be achieved with a low noise amplifier, and
installing the signal conversion equipment near the antenna. Therefore,
no analog signal processing is needed, because a Digital Signal Processor
(DSP) can do the job.
The conversion to and from the digital format is handled by
Digital-to-Analog Converters (DACs) and Analog-to-Digital Converters
(ADCs). To achieve the performance required for a cognitive radio, not
only must the DACs and ADCs have an extended dynamic range, and be
able to operate over a very wide frequency range, extending up to many
gigahertz, but in the case of the transmitter they must be able to handle
significant levels of power.

1.7 OPERATION OF COGNITIVE NETWORKS

A considerable amount of research has been done, and a number of limited

trials of cognitive radio technology have taken place, during the previous
years, but an example of a widely manufactured and distributed cognitive
radio system is still needed in the telecommunications sector.
Femtocell base stations is a promising solution to tackle the com-
munications for indoor environments. The idea of femtocells has been
developed to allow users to achieve better cell coverage and higher trans-
mission rates within their homes. Using an Asymmetric Digital Subscriber
Line (ADSL) back-haul to link into the network, turns the femtocells into
a cellular access point (Xiang et al. 2010).
The correct operation of the femtocells must not interfere with the
main network, or with any adjacent femtocells. Cognitive radio tech-
nology has been used to achieve this goal. By using cognitive radio, the
femtocells can monitor their environment, detect which geographic area
they are in to ensure they comply with regulatory standards, and then
choose a suitable channel frequency.
10 SPECTRUM SENSING TECHNIQUES AND APPLICATIONS

Table 1.1. Comparison of various indoor wireless access technologies

Specifications Femtocell Picocell Relay Wi-Fi
Typical power 10100 mW OI OI 100200 mW
Coverage range 2050 m 150 m CEM 100200 m
Services RTVD RTVD RTVD VoIP
Deployment scenarios IHO HSO HSOT IHS
Access mode COH OA OA COA
Back-haul DCO X2 WIO DCO
Peak data rate LTE-A LTE-A 802.11n

Table 1.1 shows a comparison of various indoor wireless access tech-

nologies (Huang et al. 2013). In order to understand the table, consider the
following notation:

LTE-A corresponds to LTE-Advanced (3GPP R10): 1 Gbit/s, for

Downlink (DL) and 300 Mbit/s, for Uplink (UL).
The acronym CEM corresponds to Coverage Extension of Macro.
The acronym RTVD stands for Real-Time Voice and Data.
The acronym VoIP stands for Primarily data and VoIP.
The acronym OI stands for an outdoor power of 250 mW2W, and
indoor power less than 100 mW.
The acronym IHO corresponds to In-Home or Office.
The acronym HSO stands for Hot Spot or Office.
The acronym HSOT stands for Hot Spot or office or tunnel or high-
speed train.
The acronym IHS stands for In-Home or Hot Spot.
The acronym DCO corresponds to DSL (Digital Subscriber Line)
or cable or optical fiber.
The acronym X2 stands for X2 interface.
The acronym WIO stands for Wireless In-band or Out-of-band.
The acronym COH corresponds to Closed or Open or Hybrid.
The acronym OA stands for Open Access.
The standard 802.11n corresponds to a transmission rate of 600
Mbit/s.
The acronym COA stands for Closed or Open Access.

A communication system that consists of primary networks and a

femtocell-based heterogeneous cognitive radio network is described in
(Xie et al. 2012). Figure 1.4 depicts the system model for femtocell-based
 Cognitive Radio11

PU PU
PU PU PBS
PU
PBS PBS

Bk

B1 BN

Cognitive
BS
MSU
FSU
MSU FBS

Figure 1.4. The system model for femtocell-based

cognitive radio networks. Based on (Xie et al. 2012).

cognitive radio networks, in which the following definitions are in place:

Primary User (PU), Base Station (BS), Uplink (UL), Downlink (DL),
Primary Base Station (PBS), spectrum resource from primary network k
(Bk), Macrocell Secondary User (MSU), Femtocell Base Station (FBS),
and Femtocell Secondary User (FSU).
By incorporating cognitive radio in femtocell networks, the cogni-
tive radioenabled femtocells are able to access licensed spectrum bands
not only from macrocells but also from other licensed systems, such as
TV systems. In this way, the co-channel interference in femtocells can be
reduced and the network capacity can be improved (Xiang et al. 2010).

1.8 REGULATION AND STANDARDS

Traditional regulatory structures have been built for analog transmission,

and it is not optimized for cognitive radio. Regulatory bodies, including
the FCC, in the United States, and OFCOM, in the United Kingdom, as
well as different independent measurement campaigns found that part of
the RF spectrum is inefficiently utilized.
The cellular network bands are overloaded in several cities, but other
frequency bands, such as those used for military applications, amateur
radio, and paging, are insufficiently utilized. Independent studies per-
formed in some countries confirmed that observation, and concluded that
spectrum utilization depends on time and place.
12 SPECTRUM SENSING TECHNIQUES AND APPLICATIONS

Moreover, fixed spectrum allocation prevents rarely used frequencies,

those assigned to specific services, from being utilized, even when the
unlicensed users do not cause noticeable interference with the assigned
service.
Regulatory bodies, in several countries, have been considering
whether to allow unlicensed users in licensed bands if they would not
cause any interference to licensed users. These initiatives have directed
cognitive radio research to focus on dynamic spectrum access.

1.8.1 COGNITIVE RADIO TERMINOLOGY

Depending on transmission and reception parameters, there are two main

types of cognitive radio:

Full Cognitive Radioin which all parameters observable by a

wireless node, or network, are considered.
Spectrum-Sensing Cognitive Radioin which only the RF spec-
trum is considered.

Other types are dependent on parts of the spectrum available for cog-
nitive radio:

Licensed-Band Cognitive Radiothat is capable of using bands

assigned to licensed users, except for unlicensed bands, such as the
Unlicensed National Information Infrastructure (U-NII) band or the
ISM band. The IEEE 802.22 working group develops a standard for
Wireless Regional Area Network (WRAN), to operate on unused
television channels.
Unlicensed-Band Cognitive Radiowhich is limited to unlicensed
parts of the RF spectrum. One such system is described in the IEEE
802.15 Task Group 2 specifications, which focus on the coexistence
of IEEE 802.11 and Bluetooth.
Spectrum mobilityA process by which a cognitive radio user
changes its frequency of operation. Cognitive radio networks are
designed to use the spectrum in a dynamic manner, by allowing
radio terminals to operate in the most adequate frequency band, and
maintain seamless communication requirements during transitions.
Spectrum sharingAllows cognitive radio users to share the spec-
trum bands of the licensed-band users. However, the cognitive
radio users have to restrict their transmit power so that the interfer-
ence caused to the licensed-band users is kept below a certain limit.
 Cognitive Radio13

Sensing-based spectrum sharingIn sensing-based spectrum shar-

ing cognitive radio networks, cognitive radio transceivers first
observe the spectrum allocated to the licensed users to detect the
state of the licensed users. Based on the detection results, cognitive
radio users decide their transmission strategies.
If the licensed users are not using the bands, the cognitive radio
users are allowed to transmit over those bands. If the licensed users
are using the bands, the cognitive radio users share the spectrum
bands with the licensed users by restricting their transmit power.

1.8.2 COGNITIVE RADIO FUNCTIONS

The cognitive radio technology enables several capabilities to improve the

usefulness and effectiveness of wireless communications. Those functions
include (SDR Forum 2008):
Exploit locally vacant, or unused, radio channels, or ranges of radio
spectrum, to provide new paths to spectrum access.
Roam across borders and perform self-adjustment to stay in com-
pliance with all local radio operations and emissions regulations.
Negotiate as a broker on behalf of the radio user with multiple
service providers to give a network access matched to the user
needs at the lowest cost.
Adapt itself without user intervention to save battery power or to
reduce interference to other users.
Make use of location awareness to ensure that radio emissions do
not interfere with licensed broadcasters.
Understand and follow the actions and choices taken by their users
to become more responsive and anticipate user needs over time.
Formulate and issue queries, from one radio to another.
Execute commands sent by other radios.
Combine contradictory or complementary information.
Some cognitive radio functions are important enough, to deserve sep-
arate explanations:
Power controlis generally used for spectrum sharing in cognitive
radio systems to maximize the capacity of secondary users with
interference power constraints, to protect the primary users.
Spectrum sensingthe detection of unused spectrum, to be shared,
without undue interference to other users, which is an important
requirement of the cognitive radio network, requires the equipment
to sense empty spectrum. Detecting primary users is an efficient
way to detect empty spectrum.
14 SPECTRUM SENSING TECHNIQUES AND APPLICATIONS

Spectrum-sensing techniques may be grouped into three categories:

Transmitter detectioncognitive radios must be able to determine

if a signal from a primary transmitter is locally present in a certain
spectrum.
There are several proposed approaches to transmitter detection
(SDR Forum 2008):
Matched filter detectionwhich is the optimal technique to
detect a used channel, but requires information about the trans-
mitted signal.
Energy detectionis a spectrum sensing method that detects the
presence or absence of a signal by measuring the received signal
power. This signal detection approach is convenient for practi-
cal implementation, because it only requires the noise variance
information. It has been shown that an imperfect knowledge of
the noise power (noise uncertainty) may lead to the phenomenon
of the Signal-to-Noise-Ratio (SNR) wall, which is an SNR level
below which the energy detector cannot reliably detect any trans-
mitted signal even increasing the observation time.
 Cyclostationary detectionuses spectrum sensing algorithms
based on the cyclostationary behavior of modulated signal, such
as Binary Phase Shift Keying (BPSK), Quaternary Phase Shift
Keying (QPSK), Amplitude Modulation (AM), and Orthogonal
Frequency Division Multiplex (OFDM). The noise signals do not
exhibit cyclostationary behavior, which allows the detector to dif-
ferentiate them from the modulated signals. These detectors are
robust against noise variance uncertainty.
Wideband spectrum sensingrefers to spectrum sensing over large
spectral bandwidth, typically hundreds of megahertz or even sev-
eral gigahertz. Since current ADC technology cannot afford the
high sampling rate with high resolution, it requires revolutionary
techniques such as compressive sensing and sub-Nyquist sampling.
Cooperative detectionrefers to spectrum sensing methods in
which information from multiple cognitive radio users is incor-
porated for primary user detection.
Interference-based detectionuses the interference temperature
to detect the vacant channel, which is hard to measure. It can
avoid the hidden terminal problem.
 Null-space-based cognitive radiouses multiple antennas, to
detect the null-space of the primary user and then transmit within
the null-space, so that its subsequent transmission causes less
interference to the primary user.
 Cognitive Radio15

Spectrum managementpermits to capture the best available spec-

trum to meet user communication requirements, while not creating
undue interference to the primary users. Cognitive radios should
decide on the best spectrum band to meet certain quality of service
requirements. Therefore, spectrum management functions are
required for cognitive radios.
Spectrum-management functions are classified as:
Spectrum analysisthe observation of the dynamics of the spec-
trum, as the users occupy the frequency channels.
Spectrum decisionthe result of a series of procedures to evalu-
ate the opportunity to occupy a possibly vacant channel.
The practical implementation of spectrum management func-
tions is a complex task, since it must address a set of technical
and legal requirements. An example of the former is choosing
an appropriate sensing threshold to detect other users, while the
latter is exemplified by the need to meet the rules and regulations
set out for radio spectrum access in international and national
legislation.

1.8.3 STANDARDS AND REGULATORY ISSUES

Different regulatory agencies and institutions are working in the standard-

ization and regulation of the cognitive radio technology. The main activi-
ties come from the task groups of the Standard IEEE 802.22 WRAN and
the committee IEEE Standards on Dynamic Spectrum Access Networks
(DySPAN) (Sithamparanathan and Giorgetti 2012).
The IEEE 802.22 is a standard developed for wireless regional area
networks and uses spectral holes to the allocation of the television broad-
casters frequencies (Sum et al. 2013). The development of this standard
was based on the concept of sharing of the nonutilized frequencies in
certain geographic regions. The objective is to cover areas with bad recep-
tion of television signals, rural areas, or low population density areas,
where the television broadcasters are not stimulated to invest in a suitable
infrastructure (Ko et al. 2010).
The standard was designed to frequency bands destined to analog
television signals, and dynamically operates in the spectral holes detected
in the frequency range of 54 to 852 MHz. The cognitive radios based on
this standard use OFDM with 2048 subcarriers (Sithamparanathan and
Giorgetti 2012).
The objective is to allow distinct signals such as digital and analog
TV, to operate in the same frequency range, without mutual interference,
16 SPECTRUM SENSING TECHNIQUES AND APPLICATIONS

or keeping a low level of interference on other transmissions, including

wireless microphones (Stevenson et al. 2009, Cordeiro et al. 2005).
For IEEE 802.22 WRAN, the acceptable sensing threshold is 20 dB,
and the signals must be detected even for low SNR (Lu et al. 2011, Zeng
and Liang 2009a).
The IEEE DySPAN work group was established by the IEEE Com-
munications Society and by the IEEE Electromagnetic Compatibility
Society. The main objectives are the design of systems and networks with
dynamic spectral access, and also deal with the wireless transmission
interference (Sithamparanathan and Giorgetti 2012). The IEEE DySPAN
developed a set of standards, known as IEEE 1900 (Sum et al. 2013,
Harada 2014).

1.9 SPECTRUM EFFICIENCY

The interest in cognitive radio is partially caused by the increase in effi-

ciency expected from the new paradigm of dynamic sharing of the fre-
quency spectrum. In this regard, it is useful to discuss some concepts
related to spectrum usage (SDR Forum 2008).

Efficiency metricsModulation efficiency is often expressed in

units of bits per second per hertz or bits per second per hertz per
square kilometer, because this unit of efficiency does not capture
all the effects in a RF transmission, such as the transmitted power
in the link, and also does take into account the complete ensemble
of users.
Spectrum sharingSpectrum sharing is the procedure in which
the spectrum that has been assigned to a licensed user is made
available to other users on a secondary, and noninterfering basis.
The secondary users may have arrangements with the license
holders by a cooperation agreement. The secondary users must
follow a set of specific rules, to prevent interference, while
accessing the spectrum, either with the cooperation or without
the knowledge of license holders, according to specific policies
and protocols.
Spectrum poolingSpectrum pooling refers to the act of multiple
spectrum license holders combing or pooling their spectrum allo-
cations. This requires agreements that describe means to access,
roam, and bill the spectrum usage.
 Cognitive Radio17

1.10UNDERLAY, OVERLAY, AND INTERWEAVE

Cognitive radio allows opening up the frequency bands to concurrent

operating users in a noninterfering mode. Accordingly, to make possible
spectrum sharing without causing harmful interference to existing traffic,
cognitive users should possess a minimum of information about their sur-
rounding noncognitive users. Depending on the knowledge that is needed
to coexist with the primary network, cognitive radio approaches fall into
the following classes (Goldsmith et al. 2009):

The underlay approachThe simultaneous transmission of cogni-

tive and noncognitive signals are permitted, as long as the interfer-
ence level at the primary user side remains acceptable. Exceeding
the predefined tolerable interference threshold may degrade the pri-
mary signal dramatically.
Some advanced signal processing techniques are efficient to
avoid or mitigate interference. Good examples are beam-forming
and spread spectrum (Gurjo and Alencar 1999a). Beam-forming
exploits the superposition of waves to guide the signal toward a
specific receiver using multiple antennas. In a cognitive context,
constructive or destructive interference is provoked at the intended
cognitive receiver to reduce the interference caused to noncognitive
users while focusing the signal energy in the direction of secondary
users.
Regarding the spread spectrum technique, the secondary signal
is multiplied by a spreading code to obtain a weaker signal with
wider bandwidth (Gurjo and de Alencar 1999). The resulting
spread signal causes lower interference level to noncognitive users.
The original secondary signal is recovered at the receiver side
by multiplying the input signal with the original spreading code
(Gurjo and Alencar 1999b). The concepts of spectrum overlay and
underlay are depicted in Figure 1.5.
The overlay approachIn this case, primary users share knowl-
edge of their signal characteristics and messages with the second-
ary users. In this way, the cognitive devices may enhance and assist
the licensed transmission rather than searching for spectrum access.
In order to accomplish this, the cognitive devices capture the mes-
sages sent by primary user, and use these messages to eliminate the
interference generated by the primary communication at the cog-
nitive receiver side or to improve the performance of the primary
18 SPECTRUM SENSING TECHNIQUES AND APPLICATIONS

S(w)

PU

PU

SU SU
overlay overlay
SU SU
underlay underlay

Figure 1.5. Spectrum overlay and underlay.

transmission, by relaying the accumulated messages to the primary

receiver.
This allows the cognitive transceiver to transmit at the same
time as the licensed transceiver, provided that its transmitted power
covers the energy needs of its own communication, as well as its
relaying operation. A compromise must be reached between the
improvement obtained by the relaying procedure and the interfer-
ence induced on the primary signal.
The interweave approachThis spectral coexistence approach has
the objective of allowing the signals to occupy the spectrum bands
that are vacant. The frequency spectrum must be observed to pre-
dict the state of each channel. The portions of spectrum that are
underutilized may be accessed by secondary users as long as the
primary activity remains idle.
In order to facilitate the coexistence of both primary and second-
ary traffics within the same wireless network, in an opportunistic
transmission mode, the spectrum opportunities must be periodi-
cally identified and monitored. The cognitive users may conduct
sensing operations permanently and reliably. Sensing algorithms
usually monitor and supervise the spectrum using the conventional
dimensions of frequency, time, and space.
However, other dimensions, such as the code and the angle of
arrival may be examined. On the other hand, geographic coordi-
nation by a central database, to identify the vacant gaps is also
important. Hybrid schemes, possibly using a combination of the
mentioned examples, have some potential to improve the efficiency
of spectrum sharing.
 Cognitive Radio19

In summary, the interference levels associated to the determin-

istic spectral occupancy play an important hole in the cognitive
sensing. The definition of the acceptable interference levels is fun-
damental. The observed interference in the spectral sensing can
then be classified, according to the level, as (Wyglinski et al. 2010,
Rocha 2013):
Interleaved or avoided interferenceThe secondary users use
the spectrum without interference on the primary users (or other
cognitive users). That is, the cognitive users occupy the spectral
holes without damage to the other users in operation (Biglieri et
al. 2013).
Controlled, or underlay, interferenceThe cognitive users
transmit in the same frequency band of the primary users.
However, the interference on the primary user is kept below an
acceptable level.
Mitigated interference, or overlayThe operation is in the
same channel of a primary user, but the secondary user may
have information about the licensed users in operation. This
information is useful to cancel or minimize the interference
between the primary user and the secondary user. Additionally,
it can allow the cooperation with the primary user, in order
to enhance the communication in that channel (Biglieri et al.
2013).

1.11 COGNITIVE RADIO DEFINITIONS

This section defines some concepts in the area of SDR and cognitive radio,
in line with the position of the SDR Forum with regard to these technolo-
gies. The definitions are related to wireless communications (SDR Forum
2008).

Radio Nodea wireless device point of presence incorporating a

radio transmitter or receiver.
Physical Layerthe layer within the wireless protocol in which
processing of RF, IF (Intermediate Frequency), or baseband signals
including channel coding occurs. It is the lowest layer of the ISO
7-layer model as adapted for wireless transmission and reception.
Data Link Layerthe protocol responsible for reliable frame
transmission over a wireless link through the use of adequate error
detection and control procedures and medium access control.
Software Controlledrefers to the use of software processing within
the radio system or device to select the parameters of operation.
20 SPECTRUM SENSING TECHNIQUES AND APPLICATIONS

Software Definedrefers to the use of software processing within

the radio system or device to implement operating (but not control)
functions.
Software Controlled Radio (SCR)a radio in which some or all of
the physical layer functions are software controlled.
Software Defined Radio (SDR)a radio in which some or all of the
physical layer functions are software defined.
Adaptive Radioa radio in which communications systems have
means of monitoring their own performance and a means of vary-
ing their own parameters by closed-loop action to improve their
performance.
Cognitive Radioan approach to wireless engineering wherein the
radio, radio network, or wireless system is endowed with aware-
ness, reason, and agency to intelligently adapt operational aspects
of the radio, radio network, wireless system.
The general architecture of an SDR is presented in Figure 1.6,
and is composed of (Wyglinski et al. 2010):
Antenna, which is responsible for the transmission and reception
of the digital signals.
Radiofrequency front-end, a radiofrequency transceiver for the
transmitted or received signals by the antennas (Akyildiz et al.
2006a).
Analog-to-digital and digital-to-analog converters, which p erform
conversions to allow the digital signal processing of the signals
manipulated by the SDR.

Antenna

Baseband
A/D Converter
Converter

Baseband
Radiofrequency Processing
Frontend Unity

High frequency
D/A Converter
Converter

Figure 1.6. Block diagram of a software defined radio. Based on

(Wyglinski et al. 2010).
 Cognitive Radio21

 Baseband converters (Digital Down-ConversionDDC) and

high-frequency converters (Digital Up-ConversionDUC) exe-
cute the demodulation of the received signals and the modulation
of the transmitted signals, respectively.
The baseband processing unit processes the received data. The
impact of the hardware in the development of those devices is
reduced due to the reconfigurable architecture of the SDRs
(Akyildiz et al. 2006a).
 Index

5G, 53 Bit Error Probability, 25

delivery of faster data speeds, 64 Bit error rates, 64, 83
provide wireless connectivity, 64 Black space, 5
transactions, security of, 65 Blind sensing, 30, 37
802.11p, 56, 61 detectors, 30
Bluetooth, 12, 53, 58
A
Adaptive radio (AR), definition of, C
2, 20 Carrier Sense Multiple Access
Additive White Gaussian Noise (CSMA), 44
(AWGN), 25, 3132 Coverage Extension of Macro
energy detection in, 3335 (CEM), 10
probability distribution of Centralized Cooperative Sensing,
transmitted signal, 36, 38 44
Advanced Metering Infrastructure CEPT. See European Conference
(AMI), 63 of Postal and Telecommunica
Amplitude Modulation (AM), 14 tions Administrations (CEPT)
Analog-to-Digital Converters chi-squared distribution,
(ADCs), 9, 14 7274, 76
Anderson-Darling (AD), 37 Cluster-Based Cooperative
And Fusion Rule, 48 Spectrum Sensing, 49
Antenna, 9, 20 Cognitive
multiple, 14, 17, 46 architecture, 69
single CR terminals, 43 functions, 1315
Architecture interweave, 1719
cognitive network, 69 overlay, 1719
cognitive WSN, 6668 radio, 121, 23, 3031, 4147,
CRV network, 5960 49, 5354, 5766, 68, 73
information process, 78 full, 4, 12
licensed-band, 12
B spectrum sensing, 4, 12
Base station (BS), 9, 11, 59 unlicensed-band, 12
Binary Phase Shift Keying underlay, 1719
(BPSK), 14 user, 2
102Index

Cognitive Base Transceiver Digital Signal Processor (DSP), 9

Station (CBTS), 44 Digital TV, 2, 54
Cognitive Networks, 4, 69, 35, Digital Up-Conversion (DUC), 21
42, 45, 48, 58, 64, 67, 69 Distributed Spectrum Sensing,
Cognitive Radio Vehicular ad hoc 5052
NETworks (CR-VANET), 54, Distribution
5859, 62 chi-squared, 39, 7273
Cognitive Radio-Wireless Sensor exponential, 69
Networks (CR-WSNs), 65 gamma, 7374
Cognitive Vehicular Networks, 53, Gauss, 3639, 72, 76, 78, 8081
58, 6162 lognormal, 7778
Constant Detection Rate (CDR), Nakagami, 77
34 Pareto, 8182
Constant False Alarm Rate probability, 25, 29, 31, 34,
(CFAR), 34 3638, 46, 73, 78, 80, 82
Control power, 13 Rayleigh, 7475
Control Channel, 42, 4445, 47, Rice, 7577
49 von Mises, 7880
Cooperation, 16, 19, 4145, 52, 58 Weibull, 8284
Cooperative detection, 14, 48, 50 Downlink (DL), 1011
Cooperative Spectrum Sensing, Driving Safety Support Systems
41, 4350, 52 (DSSS), 62
Covariance, 31 Dynamic Spectrum Access
Cyclostationary detection, 14, 31, Networks (DySPAN), 1516
58 Dynamic Spectrum Allocation
(DSA), 2
D
Data Fusion, 45 E
Data link layer, 19 Efficiency
Decision Rule, 2526, 46 metrics, 16
Dedicated Short-Range spectrum, 16
Communications (DSRC), Electronic Communications
5657, 62 Committee (ECC), 54
Degree of Freedom, 34, 38, 72 Emergency Networks, 53
Detection Empirical Cumulative Distribution
cooperative, 14 Function (ECDF), 38
cyclostationary, 14 Energy
energy, 14 detection, 14, 3035, 58
interference-based, 14 detector, 14
matched filter, 14 efficient sensor nodes, 66
probability, 2729, 31, 3435, 49 electrical, 62
transmitter, 14 harvesting, 63
Digital-to-Analog Converter signal, 17
(DAC), 9, 20 Equal Gain Soft Decision Fusion,
Digital Down-Conversion (DDC), 46
21 Estimation, 2829
 Index 103

European Conference of Postal Generalized Likelihood-Ratio Test

and Telecommunications (GLRT), 29
Administrations (CEPT), 54 Goodness of Fit (GoF), 36
European Union, 2 Goodness of fit Higher Order
Expected Value, 36, 76, 80, 82 Statistics Testing (GHOST), 39
Gray space, 5
F
Fading, 35, 4142, 49, 57, 69, 76 H
False Alarm Probability, 28, Hard Combining, 45, 4749, 51
3435, 49 Health Care Networks, 64
Fast Fourier Transform (FFT), Heterogeneous
3233, 3839 cognitive radio network, 10
Federal Communications multiple sensor networks, 68
Commission (FCC), 12, 11, network, 10
5354 Hidden Terminal, 14, 4142, 66
Federal Office of Communications Home Area Networks (HAN), 63
(OFCOM), 1, 11, 5354 Hybrid Spectrum Sensing, 31
Femtocell, 10 Hypothesis Test, 2425, 36,
networks, 11 4546, 69, 73
radio-enabled, 11
Femtocell base stations (FBS), 9, I
11 Infotainment, 5758
Femtocell Secondary User (FSU), Intelligent radio (IR), 23
11 Intelligent Transportation Systems
Frequency (ITS), 55
bands, 12, 46, 11, 17, 2324, Interference-based detection, 14
27, 58, 65, 67 Internet of Things (IoT), 6465
carrier, 31 Interweave, 1719
channels, 1, 15 ITU-T. See Telecommunication
domain, 32 Standardization Sector (ITU-T)
spectrum, 2, 16, 18, 53
Front-end, 9, 20, 44 J
Function Jarque-Bera (JB), 3739
Gamma, 35, 73, 77
Fusion Center (FC), 4244, 50, K
58, 61 Kolmogorov-Smirnov (KS), 37
Fusion Rules, 43, 46, 48 Kurtosis, 3739

G L
Gamma Language modeling, 8
distribution, 7374 Laplace, distribution, 6971
function, 35, 77 Laplace, Pierre-Simon de, 69
Gaussian Likelihood Ratio Test, 29
distribution, 3639, 72, 76, 78, Likelihood Test (LRT), 29, 46
8081 Lilliefors, 3738
noise, 25, 38, 71 Log-Likelihood Test, 48
104Index

Lognormal distribution, 7778 O

LTE-Advanced, 10 On-Board Unit (OBU), 56, 59
Open Systems Interconnection
M (OSI), 7
Majority Fusion Rule, 48 Optimal Fusion Rule, 46
Matched filter detection, 14, 30, Or Fusion Rule, 48
58 Orthogonal Frequency Division
Maximal Ratio Soft Decision Multiplexing (OFDM), 1415,
Fusion, 46 29
Maximum-Likelihood (ML), 29 Overlay, 1719
Mean, 20, 31, 3638, 62, 6972,
7477, 79 P
Medical Body Area Networks Parallel Distributed Sensing, 51
(MBANs), 64 Parametric Tests, 37
Medium Access Control (MAC) Pareto, Vilfredo Federico Damaso,
layer, 4, 19 8182
Missed Detection Probability, Pearson Chi-square, 37
2829, 49, 52 Physical layer (PHY), 4, 7, 1920
Mobile Ad Hoc Network Power, 34, 9, 31
(MANET), 55 consumption, 30
Multipath, 41, 7576 control, 13
fading, 57, 77 grid, 63
propagation, 79 interference, high and low, 5
scatter components, 84 line communications, 53, 6263
Multitaper, 31 noise, 14, 30, 35
operators, 63
N spectrum, 5
National Telecommunications substation, 63
Agency (Anatel), 1 supply, 61
N-out-of-M Fusion, 48 transmitted, 16, 18
Network, 12 Power Line Communications
heterogeneous, 10 (PLC), 53, 6263
primary, 10 Primary Users (PU), 2, 4, 6, 11,
Neyman-Pearson (NP), 2829 1315, 17, 19, 2324, 3033,
Noise 3638, 41, 4351, 62
amplifier, 9 Probability Distribution Function,
environmental, 5 29, 31, 34, 36, 46, 80
Gaussian, 38, 71
in analog and digital Q
communications, 71 Quality of Service (QoS), 3, 15,
uncertainty, 14 60, 64
variance, 14, 29 Quantization, 47, 87
Non-Blind Sensing, 30 Quantized Soft Combining, 47
Non-Parametric Tests, 37 Quaternary Phase Shift Keying
Null Hypothesis, 2526 (QPSK), 14
 Index 105

R Signal fading, 69
Radio Signal to Noise Ratio (SNR), 14,
adaptive, 2, 20 16, 27, 32, 35, 4647, 67
cognitive, 121, 23, 3031, Significance Level, 26
4147, 49, 5354, 5766, 68, Skewness, 3639
73 Smart Grids, 53, 6263
environment map, 8 Soft Combining, 4547, 49
FM, frequency of, 1 Software, 2
intelligent, 23 benefits of, 6
node, 19 controlled, 19
Random Variable (RV), 25, 34, controlled radio, 20
3637, 71, 7980, 8284 defined, 20
Rayleigh distribution, 7475, 77, Software-Defined Radio (SDR),
83 24, 67, 1314, 1921, 58
Real-Time Voice and Data Spatial Diversity, 4142
(RTVD), 10 Spectral
Receiver Operating Characteristic allocation, 66
(ROC), 28 bands, 66
Rice distribution, 7577 bandwidth, 14
Roadside Unit (RSU), 56, 59 decision parameters, 67
density, 31
S detection, 5, 23, 27
Secondary Users (SU), 2, 4, 6, 13, detector, 26
1619, 23-24, 31, 41, 44, 48, handoff, 6
5051 holes, 46, 15
Semi-Blind Sensing, 30 management, 6
Sensing mobility, 6
algorithms, 18 occupancy, 2, 5, 19
blind, 30, 37 sensing, 4, 6, 8, 10, 1314,
circuitry, 65 2325, 2733, 3639, 4145,
cooperative, 49 49, 54, 5859, 61, 63, 67, 69
distributed spectrum, 50 sharing, 6
indoor applications, 68 Spectral Detection, 5, 23, 27
spectrum, 4, 6, 8, 10, 1314, Spectrum
2325, 2733, 3639, 4145, analysis, 15
49, 54, 5859, 61, 63, 67, 69 band, 1, 1113, 15, 18, 64
station, 44 decision, 15
threshold, 32 efficiency, 16
time, 42 hole, 23, 5, 23, 3031, 38, 41,
wideband spectrum, 14 58, 61, 69
Sequential Distributed Sensing, management, 15
5152 mobility, 12, 67
Shapiro-Wilk (SW), 37 pooling, 16
Signal Detection, 14, 2829, 31, regulations, 1116
47 sensing, 4, 6, 8, 10, 1314,
106Index

2325, 2733, 3639, 4145, V

49, 54, 5859, 61, 63, 67, 69 Variance, 3132, 34, 3637, 71,
sensing-based, 13 7475, 77, 82
sharing, 1213, 1618, 54, 67 for unitary mean, 76
Spectrum Sensing, 4, 6, 8, 10, 1314, noise, 14, 2930
2325, 2733, 3639, 4145, 49, signal, 29, 38
54, 5859, 61, 63, 67, 69 Vehicle-to-Infrastructure (V2I)
Station base, 9, 11, 42, 44, 56, 59 communication, 5657
Statistical Test, 28, 3132, 34, Vehicle-to-Person (V2P)
3639 communication, 5657, 59, 62
Vehicle-to-Roadside (V2R), 56
T Vehicle-to-Vehicle (V2V), 5557,
Telecommunication 59, 62
Standardization Sector Vehicular Networks, 53, 5759,
(ITU-T), 7 6162
Test and measurement, 8 Voice over Internet Protocol
Threshold, 17, 28, 30, 3234, (VoIP), 10
3839, 46, 48 von Mises distribution, 7881
Transmitter, 9, 24, 27, 4344, 50, 75 Very High Frequency (VHF), 2
detection, 14
primary, 23 W
radio, 19 Weibull, distribution, 8284
TV White Spaces, 53, 61, 64 Weibull, exponentiated
Type I Error, 26 distribution, 8284
Type II Error, 26 Weibull, Waloddi, 82
White space, 5, 53, 6365
U bands, 54
Uplink (UL), 1011 database, 54
Ultra High Frequency (UHF), 2, Wide Area Networks (WAN), 63
5354 Wireless Access in Vehicle
Underlay, 1718 Environment (WAVE), 5556,
USA, 2, 11, 5354, 56 61
User, 34 Wireless Fidelity (WiFi), 10, 53
cognitive, 2, 56 Wireless Innovation Forum (WIF),
licensed, 6, 1213 7
network, 6 Wireless Local Area Network
noncognitive, 17 (WLAN), 44
primary, 2, 4, 6, 11, 13, 15 Wireless Regional Area Network
radio, 2 (WRAN), 12, 1516
secondary, 2, 4, 6, 11, 16, 23 Wireless Sensor Network (WSN),
unlicensed, 12 4, 55, 6568