Performance of Cooperative Spectrum Sensing in Cognitive Radio Networks

TABLE OF CONTENTS

ACKNOWLEDGEMENTS

ABSTRACT

LIST OF FIGURES

LIST OF TABLES

ABBREVIATIONS

NOTATIONS

1 INTRODUCTION
1.1 Characteristics of Cognitive Radio
1.1.1 Cognitive Capability
1.1.2 Reconfigurability
1.2 Architecture of Cognitive Radio Network
1.2.1 Primary Network
1.2.2 Cognitive Radio Network
1.3 Cognitive Cycle
1.3.1 Spectrum Sensing
1.3.2 Spectrum Decision
1.3.3 Spectrum Sharing
1.3.4 Spectrum Mobility
1.4 Classification of Spectrum Sensing Techniques
1.4.1 Primary Transmitter Detection
1.4.2 Receiver Detection
1.4.3 Interference Temperature Management
1.5 Evaluation Metrics
1.6 Conventional versus Cooperative Spectrum Sensing
1.7 Cooperative Spectrum Sensing
1.7.1 Centralized Cooperative Spectrum Sensing
1.7.2 Distributed Cooperative Spectrum Sensing
1.7.3 Relay Assisted Cooperative Spectrum Sensing
1.8 Elements of Cooperative Spectrum Sensing
1.8.1 Cooperation Models
1.8.2 Sensing Techniques
1.8.3 Hypothesis Testing
1.8.4 Control Channel and Reporting
1.8.5 Data Fusion
1.8.6 User Selection
1.8.7 Knowledge Base
1.9 Framework of Cooperative Spectrum Sensing
1.10 Research Objectives
1.10.1 General Objectives
1.10.2 Specific Objectives

2 REVIEW OF LITERATURE
2.1 Background
2.2 Review of Cognitive Radio Networks
2.3 Review of Cooperative Spectrum Sensing
2.4 Review of Energy Efficiency in Cooperative Spectrum Sensing
2.5 Review of Optimization in Cooperative Spectrum Sensing
2.6 Review of Cooperative Spectrum Sensing Under Fading Channels
2.7 Research Gap
2.4 Summary of Literature Review

3 OPTIMIZATION OF DECISION THRESHOLD IN COOPERATIVE SPECTRUM SENISNG
3.1 Introduction
3.2 System Model
3.3 Analysis of Energy Efficiency
3.4 Optimization of Final Decision Threshold
3.5 Conclusion

4 ENERGY EFFICIENCY IN COGNITIVE RADIO NETWORKS USING COOPERATIVE SPECTRUM SENSING
4.1 Introduction
4.2 System Model
4.3 Average Channel Throughput of OR, AND and MAJORITY Fusion Rules
4.4 Energy Consumption of OR, AND and MAJORITY Fusion Rules
4.5 Energy Efficiency of OR, AND and MAJORITY Fusion Rules
4.6 Conclusion

5 OPTIMIZATION OF COOPERATIVE SECONDARY USERS IN COGNITIVE RADIO NETWORKS
5.1 Introduction
5.2 Optimization of Cooperative Secondary Users using OR, AND Fusion Rules
5.2.1 Optimal Number of Cooperative Secondary Users using OR Fusion Rule
5.2.2 Optimal Number of Cooperative Secondary Users using AND Fusion Rule
5.3 Performance of OR, AND Fusion Rules
5.4 Conclusion

6 PERFORMANCE OF COOPERATIVE SPECTRUM SENSING OVER RAYLEIGH FADING CHANNELS
6.1 Introduction
6.2 Probability of Detection and Probability of False Alarm in Cooperative Spectrum Sensing
6.3 Total Error Rate
6.4 Success Probability
6.5 Conclusion

7 CONCLUSION AND FUTURE SCOPE

REFERENCES

LIST OF PUBLICATIONS

DECLARATION

I hereby declare that the thesis entitled “PERFORMANCE OF COOPERATIVE SPECTRUM SENSING IN COGNITIVE RADIO NETWORKS”, submitted to the Department of Electronics and Communication Engineering, Acharya Nagarjuna University, in partial fulfillment of the requirement for the award of the Degree of Doctor of Philosophy is a bonafide record of my research work carried out under the supervision of Dr. Manchikalapudi Satya Sai Ram, Associate Professor, Department of Electronics and Communication Engineering, RVR & JC College of Engineering, Chowdavaram, Guntur. I further declare that the thesis has not been submitted earlier by me or others in part or full for the award of any Degree in any University.

CERTIFICATE

This is to certify that the present dissertation work entitled, “PERFORMANCE OF COOPERATIVE SPECTRUM SENSING IN COGNITIVE RADIO NETWORKS”, submitted by Mrs. CHILAKALA SUDHAMANI for the award of the Degree of Doctor of Philosophy in Electronics and Communication Engineering to the Acharya Nagarjuna University, Nagarjuna Nagar, is a record of bonafide research work carried out under my supervision in the Department of Electronics and Communication Engineering of this University. The results embodied in this dissertation have not been submitted for any degree or diploma.

CERTIFICATE

This is to certify that all corrections and suggestions pointed out by the State/Indian/Foreign Examiner(s) are incorporated in the Thesis titled “PERFORMANCE OF COOPERATIVE SPECTRUM SENSING IN COGNITIVE RADIO NETWORKS”, submitted by Mrs. CHILAKALA SUDHAMANI.

Thesis Title: “Performance of Cooperative Spectrum Sensing in Cognitive Radio Networks”

DETAILED CORRECTION REPORT

Abbildung in dieser Leseprobe nicht enthalten

ACHARYA NAGARJUNA UNIVERSITY NAGARJUNA NAGAR

Declaration for Plagiarism Check

It is certified that the Ph.D. Thesis / M.Phil. dissertation titled “PERFORMANCE OF COOPERATIVE SPECTRUM SENSING IN COGNITIVE RADIO NETWORKS” by Mrs. CHILAKALA SUDHAMANI has been examined by us.

We undertake as follows:

a. Thesis has significant new work/knowledge as compared to already published or are under consideration to be published elsewhere. No sentence, equation, diagram, table, paragraph or section has been copied verbatim from previous work unless it is placed under quotation marks and duly referenced.
b. The work presented is original and own work of the author (i.e. there is no plagiarism). No ideas, processes, results or words of others have been presented as Authors own work.
c. There is no fabrication of data or results which have been compiled/analyzed.
d. There is no falsification by manipulating research materials, equipment or processes, or changing or omitting data or results such that the research is not accurately represented in the research record.
e. The thesis has been checked using TURNITIN.

ACKNOWLEDGEMENTS

I thank God for his blessings in pursuing Ph.D, which has been truly challenging and unforgettable experience. This great accomplishment is only possible through the support and guidance from many people.

Firstly, I want to stretch out gratitude to my research supervisor Dr. M. Satya Sai Ram, Associate Professor, Department of ECE, RVR & JC College of Engineering, Chowdavaram Guntur, for all the support and encouragement given to me. I thank him for sowing the initial seeds of Wireless Communication and MATLAB, which has created great interest in me towards this area. I thank him for making me realize the importance of research contribution in science citation journals.

Many thanks to Dr. P. Siddaiah, Principal, Faculty of Engineering, ANU College of Engineering and Technology, Acharya Nagarjuna University, for helping me to do my research work. My special thanks to Dr. E. Srinivasa Reddy, Dean, ANU College of Engineering and Technology, ANU, who supported me in completing my work. I thank the ECE Department of ANU for organizing the reviews and furnishing valuable suggestions as well as providing the necessary tools and infrastructure to work for my Ph.D.

I thank my colleagues Dr. Ashutosh Saxena, Dr. A. Raji Reddy, G. Srikanth, B. Surekha, B. Narasimha, V. Aswini, Suraya Mubeen, P. Sireesha, Y. Priyanka, D. Sreekanth, M. Geetha, N. Renuka who are always there to support me throughout the perusal of my Ph.D.

I extend thanks to my parents Mr. Ch. Bali Reddy, Mrs. Ch. Obulamma and parents-in-law Mrs. Ch. Pulla reddy, Ch. Shivanagendramma for always inspiring me to achieve higher qualification. Their blessings have been a valuable asset to me. I thank from the core of my heart my husband Ch. Venkata Mohan Reddy and my daughter Ch. Siva Sai Sri Varsha Reddy for their enthusiastic help, constructive suggestions, patience and understanding.

Last but not the least, I would like to acknowledge every person who has been involved directly or indirectly with my doctoral study and for having shown much care and consideration as I spent the hours toiling away on this thesis.

ABSTRACT

The rapid growth in wireless technology undoubtedly brings an increasing need of spectrum resources. Federal communication commission is an independent agency used for managing and licensing the electromagnetic spectrum. It identifies the under utilized spectrum by the licensed users or primary users. Maximizing the utilization of radio spectrum has become a major concern topic of research. Hence cognitive radio has been proposed as a new technology for maximizing the spectrum utilization. It enables the secondary users to sense and detect the presence or absence of the primary user. So that the free spectrum can be utilized by the secondary users when the primary user is absent, without causing any interference to the primary user. Detecting the primary users using single secondary user is not reliable due to multipath fading, shadowing and receiver uncertainty i.e. lack of knowledge and ability to model and measure the channel. These problems may cause secondary users to access the licensed spectrum and it cause interference to the primary users. Therefore to overcome these issues and to enhance the detection accuracy, cooperative spectrum sensing technique has been proposed. The basic idea of cooperative spectrum sensing is to improve the sensing performance, reliability, accuracy, reducing the sensing time and the interference caused by secondary users by allowing the cooperation among the secondary users.

Cooperative spectrum sensing technique is used to maximize the utilization of unused licensed spectrum and it demonstrates the cooperation among the secondary users. As the cooperation among the secondary users increases the detection performance increases, which increases the average channel throughput and energy efficiency but it depends on the number of cooperative secondary users, fusion rules, channel conditions and detection threshold. In this thesis average channel throughput, energy consumption and energy efficiency are estimated for variable number of secondary users and detection thresholds using hard fusion rules i.e. AND, OR and MAJORITY fusion rules. From the results it has been observed that the performance of AND fusion rule is better at low detection thresholds and for less number of secondary users. The performance of OR fusion rule is better at high detection thresholds and for large number of secondary users. The performance of MAJORITY fusion rule follows the performance of AND fusion rule at low detection thresholds and it follows the performance of OR fusion rule at high detection thresholds.

However as the number of cooperative secondary users increases the energy required for spectrum sensing and reporting sensing results to the fusion center increases, which increases the energy consumption and reduces the energy efficiency. Therefore energy efficiency can be improved by maximizing the average channel throughput or by minimizing the energy consumption. To minimize the energy consumption in cooperative spectrum sensing, optimization technique has been proposed in this thesis and it is used for further improvement of energy efficiency. With this optimization technique, optimal number of cooperative secondary users are derived by maximizing the energy efficiency using AND and OR fusion rules but not with MAJORITY fusion rule. Because it is very difficult to estimate the optimal number of cooperative secondary users using MAJORITY fusion rule, so optimization of final decision threshold was proposed in the existing methods to maximize the energy efficiency using MAJORITY fusion rule. Therefore AND and OR fusion rules are used in this work to optimize the number of cooperative secondary users. The performance of AND fusion rule is compared with the performance of OR fusion rule and it is observed that the optimal number of cooperative secondary users increases with the detection threshold in the OR rule and it decreases with the detection threshold in the AND rule.

Apart from considering a cooperative spectrum sensing technique for optimization of cooperative secondary users in cognitive radio networks, this work also considered a cooperative spectrum sensing with imperfect reporting channels. Cooperative spectrum sensing with imperfect reporting channels between secondary users and fusion center is proposed and estimated the total error rate and success probability for variable number of cooperative secondary users for fading and non fading channels. An AWGN channel is a basic noise channel and it does not include fading, interference, non-linearity and dispersion. This non fading AWGN channel and Rayleigh fading channels are considered to estimate the total error rate and success probability. From simulation results, it is observed that the number of secondary users used to maximize the success probability and to reduce the total error rate is L = 5 under both fading and non fading channels.

LIST OF FIGURES

Figure 1.1 Cognitive Radio Network Architecture

Figure 1.2 Cognitive Cycle of Cognitive Radio

Figure 1.3 Spectrum Hole Concept

Figure 1.4 Spectrum Sensing Techniques

Figure 1.5 Block Diagram of Matched Filter Detection

Figure 1.6 Block Diagram of Energy Detector

Figure 1.7 Block Diagram of Cyclostationary Future Detector

Figure 1.8 Block Diagram of Super Heterodyne Receiver

Figure 1.9 Sensor Nodes in Cognitive Radio

Figure 1.10 Interference Temperature Model

Figure 1.11 Multipath Fading, Shadowing and Receiver Uncertainty

Figure 1.12 Classification of Cooperative Spectrum Sensing, (a) Centralized, (b) Distributed and (c) Relay Assisted

Figure 1.13 Elements of Cooperative Spectrum Sensing

Figure 1.14 Parallel Fusion model

Figure 1.15 Knowledge Base Cooperative Spectrum Sensing

Figure 1.16 Framework of Cooperative Spectrum Sensing

Figure 3.1 System Model of Cooperative Spectrum Sensing

Figure 3.2 Energy Efficiency with Detection Threshold for L=1,2,3,4,5,10,20,50 under AWGN Channel

Figure 3.3 Optimal Detection Threshold with Number of Cooperative Secondary Users under Different Channels

Figure 3.4 Optimal L with Detection Threshold for SNR=10dB, 20dB, 30dB

Figure 3.5 Energy Efficiency with Detection Threshold for Rayleigh Fading Channel under Different Sensing Strategies

Figure 3.6 Energy Efficiency with Detection Threshold for Different Fading Channel under Various Fusion Rules

Figure 3.7 Energy Efficiency with Detection Threshold for Different P(H0) Values in Rayleigh Fading Channels

Figure 4.1 System Model for Cooperative Spectrum Sensing

Figure 4.2 Channel Throughput versus Detection Threshold for L=1,2,3,4,5,10,20,50 using OR Fusion Rule

Figure 4.3 Channel Throughput versus Detection Threshold for L=1,2,3,4,5,10,20,50 using AND Fusion Rule

Figure 4.4 Channel Throughput versus Detection Threshold for L=1,2,3,4,5,10,20,50 using MAJORITY Fusion Rule

Figure 4.5 Energy Consumption with Detection Threshold for L=1,2,3,4,5,10,20,50 using OR Fusion Rule

Figure 4.6 Energy Consumption versus Detection Threshold for L=1,2,3,4,5,10,20,50 using AND Fusion Rule

Figure 4.7 Energy Consumption versus Detection Threshold for L=1,2,3,4,5,10,20,50 using MAJORITY Fusion Rule

Figure 4.8 Energy Efficiency versus Detection Threshold for L=1,2,3,4,5,10,20,50 using OR Fusion Rule

Figure 4.9 Energy Efficiency versus Detection Threshold for L=1,2,3,4,5,10,20,50 using AND Fusion Rule

Figure 4.10 Energy Efficiency versus Detection Threshold for L=1,2,3,4,5,10,20,50 using MAJORITY Fusion Rule

Figure 4.11 Energy Efficiency versus Detection Threshold for OR, AND, MAJORITY Fusion Rules

Figure 5.1 Energy Efficiency versus Cooperative Secondary Users with Detection Threshold using OR Fusion Rule

Figure 5.2 Energy Efficiency versus Cooperative Secondary Users with SNR using OR Fusion Rule

Figure 5.3 Optimal Number of Cooperative Secondary Users with Detection Threshold using OR Fusion Rule

Figure 5.4 Energy Efficiency versus Cooperative Secondary Users with Detection Threshold in AND Fusion Rule

Figure 5.5 Energy Efficiency versus Cooperative Secondary Users with SNR in AND Fusion Rule

Figure 5.6 Optimal Number of Cooperative Secondary Users with Detection Threshold using AND Fusion Rule

Figure 5.7 Energy Efficiency for AND, OR Fusion Rules

Figure 6.1 Cooperative Spectrum Sensing with Imperfect Reporting Channels

Figure 6.2 Total Error Rate for AND, OR and MAJORITY Fusion Rules with Cooperative Secondary Users

Figure 6.3 Total Error Rate for AND, OR and MAJORITY Fusion Rules with Detection Threshold

Figure 6.4 Total Error Rate with Detection Threshold for L=1,2,3,4,5,6,7,8,9,10

Figure 6.5 Minimum Error Rate with Cooperative Secondary Users

Figure 6.6 Success Probability with Detection Threshold for L = 1

Figure 6.7 Success Probability with Detection Threshold for L = 5

Figure 6.8 Success Probability with Detection Threshold for L = 10

Figure 6.9 Success Probability with Detection Threshold for L = 20

Figure 6.10 Success Probability with SNR (dB) for L = 1,2,3,4,5,6,7,8,9,10

Figure 6.11 Success Probability with Number of Cooperative Secondary Users for Different SNR Values

LIST OF TABLES

Table 3.1 Energy Efficiency with Detection Threshold for L=1,2,3,4,5,10,20,50 under AWGN Channel using MAJORITY Fusion Rule

Table 3.2 Optimal Detection Threshold with Number of Cooperative Secondary Users under Different Channels

Table 3.3 Optimal L with Detection Threshold for SNR=10dB, 20dB, 30dB

Table 3.4 Energy Efficiency with Detection Threshold for Rayleigh Fading Channel under Different Sensing Strategies

Table 3.5 Energy Efficiency with Detection Threshold for Different Fading Channel under Various Fusion Rules

Table 3.6 Energy Efficiency with Detection Threshold for Different P(H0) Values in Rayleigh Fading Channels

Table 4.1 Average Channel Throughput with Detection Threshold for L=1,2,3,4,5,10,20,50 using OR Fusion Rule

Table 4.2 Average Channel Throughput with Detection Threshold for L=1,2,3,4,5,10,20,50 using AND Fusion Rule

Table 4.3 Average Channel Throughput with Detection Threshold for L=1,2,3,4,5,10,20,50 using MAJORITY Fusion Rule

Table 4.4 Energy Consumption with Detection Threshold for L=1,2,3,4,5,10,20,50 using OR Fusion Rule

Table 4.5 Energy Consumption with Detection Threshold for L=1,2,3,4,5,10,20,50 using AND Fusion Rule

Table 4.6 Energy Consumption with Detection Threshold for L=1,2,3,4,5,10,20,50 using MAJORITY Fusion Rule

Table 4.7 Energy Efficiency versus Detection Threshold for L=1,2,3,4,5,10,20,50 using OR Fusion Rule

Table 4.8 Energy Efficiency versus Detection Threshold for L=1,2,3,4,5,10,20,50 using AND Fusion Rule

Table 4.9 Energy Efficiency versus Detection Threshold for L=1,2,3,4,5,10,20,50 using MAJORITY Fusion Rule

Table 5.1 Optimal Number of Cooperative Secondary Users with Detection Threshold using OR Fusion Rule

Table 5.2 Optimal Number of Cooperative Secondary Users with Detection Threshold using AND Fusion Rule

Table 6.1 Total Error Rate with Detection Threshold for L=1,2,3,4,5,6,7,8,9,10

Table 6.2 Minimum Error Rate with Cooperative Secondary Users

Table 6.3 Success Probability with Detection Threshold for L = 1 and L = 5

Table 6.4 Success Probability with Detection Threshold for L = 10 and L = 20

ABBREVIATIONS

Abbildung in dieser Leseprobe nicht enthalten

CHAPTER 1

INTRODUCTION

The Federal Communications Commission (FCC) is responsible for regulation of interstate telecommunication and licensing of electromagnetic spectrum within the United States. It is responsible for managing and licensing the electromagnetic spectrum for commercial users and for non-commercial users including fixed and mobile wireless services, broadcast television and radio, satellite and other services. In India, Telecom Regulatory Authority of India (TRAI) will provide spectrum licenses to the mobile operators. It regularly issues orders and directions on various subjects such as tariffs, interconnections, quality of service and mobile number portability. A few unlicensed bands were left open for anyone to use as long as they followed certain power regulations. With the recent increase in wireless technologies, these unlicensed bands have become congested with everything from wireless networks to digital cordless phones.

To overcome this problem, FCC has been identifying the new ways to manage spectrum resources. The basic idea of FCC is to allow unlicensed users to use licensed band of frequencies and they should never cause interference to the primary users. Therefore Cognitive Radio (CR) has been proposed to do this practically. This can smartly sense and adapt with the changing environment by adjusting its transmitting parameters like operating frequency, modulation, frame format etc.

In the early days of communication fixed radios were used to fix the transmitter parameters. The new era of communication includes Software Defined Radio (SDR). SDR is a radio that contains a transmitter with the operating parameters like frequency range, modulation type and radiated power. These parameters can be varied by making changes in the software without changing the hardware 1. SDR is used to minimize the hardware requirements, which gives cheaper and reliable solutions to the user. But it will not take spectrum availability into account. Hence cognitive radio has been proposed as an alternative to SDR. In CR, the transmitter parameters are varied like SDR but it change its parameters intelligently according to the spectrum availability.

With in the 6 GHz spectrum band, only 0-2 GHz was effectively utilized, 2-3 GHz was moderately utilized and 3-6 GHz was utilized very less by the primary users 2.

However to improve the spectrum efficiency dynamic spectrum access techniques are proposed. The dynamic spectrum access techniques will allow the cognitive radio to operate in the best available channel. Practically the cognitive radio technology will enable the Cognitive Radio Users (CRs) or Secondary Users (SUs) to identify the underutilized portions of licensed spectrum, to detect the presence or absence of the Primary User (PU) (spectrum sensing), to select the best available channel (spectrum management), to coordinate the access of the channel with other users (spectrum sharing) and migrating to some other channel whenever the primary user is detected (spectrum mobility) 3.

1.1 Characteristics of Cognitive Radio

Cognitive radio dynamically chooses the frequency of operation and also change its transmitter parameters according to the basic characteristics of operating frequency. The basic characteristics of cognitive radios are Cognitive Capability and Reconfigurability.

1.1.1 Cognitive Capability

Cognitive capability refers to the ability of a radio to sense the information from its environment and perform real time interaction. This can be explained with three parameters. They are Spectrum Sensing, Spectrum Analysis and Spectrum Decision. Spectrum sensing performs the task of monitoring and detecting the spectrum holes and spectrum analysis will evaluate the characteristics of identified spectrum holes. In the spectrum decision, the required spectrum is selected based on the parameters like data rate, modulation type and transmission mode etc.

1.1.2 Reconfigurability

Reconfigurability refers to the ability of a radio that allows the cognitive radio to adjust its parameters like communication link, operating frequency, modulation and transmission power at run time without any modifications in the hardware components.

1.2 Architecture of Cognitive Radio Network

A clear representation of cognitive radio architecture is shown in the Figure 1.1, where some portion of the spectrum is licensed and some is unlicensed (ISM band)3.

Abbildung in dieser Leseprobe nicht enthalten

Figure 1.1: Cognitive Radio Network Architecture

Other cognitive radio networks

There are two types of networks in the architecture of cognitive radio networks. They are primary network and secondary network, which are defined as follows

1.2.1 Primary Network

Primary network is defined as the network with license to utilize a specific spectrum band. The TV broadcast network, CDMA and common cellular network are some of the examples of primary network. This network requires some basic components like primary user and primary base station.

Primary User

The user of the primary network who is having license to use certain frequency spectrum is known as primary user. The primary user will access the network through primary base station and all of its services are also controlled by the primary base station. Hence it should not be affected by any unlicensed user or any other network user.

Primary Base Station

The primary base station contains fixed network components for a required technology with licensed band. This primary network does not have capability to coexist with the cognitive radio network. Therefore it should require some modifications to access both licensed and unlicensed or cognitive radio protocols.

1.2.2 Cognitive Radio Network

Cognitive radio network (CRN) does not have license to use a desired band of frequency spectrum but it can access the licensed spectrum when it is underutilized by the primary user without causing any interference. It can be deployed as an infrastructure network or an adhoc network, which is shown in the Figure 1.1. The CR network contains some basic components, which are cognitive radio user, cognitive radio base station and spectrum broker.

Cognitive Radio User

Cognitive radio user has no license to access a fixed frequency band for its operation. Therefore cognitive radio user need some additional functionality to utilize the primary users spectrum bands.

Non cognitive users and cognitive users can be differentiated based on the priority of utilization and the legal rights on the usage of a licensed spectrum. Non cognitive users has highest priority and also has legal rights on the usage of a specific part of the electromagnetic spectrum where as cognitive users has lowest priority to utilize the primary user’s licensed spectrum and has no license on any band of spectrum.

Cognitive Radio Base station

Cognitive radio base station or secondary base station is a fixed infrastructure and it provides single hop connection between CR users without any license of a frequency spectrum. Any cognitive radio user can access the other networks with the help of this connection.

Spectrum Broker

Spectrum broker will distribute spectrum resources among different cognitive radio networks and it will make sure a fair spectrum split and separation. When different secondary users are trying to access spectrum resources then they can make a request to the spectrum broker. If free spectrum is available, then it will be assigned to the requested user to access. Therefore spectrum broker can be connected to each network in a star topology and it will acts as a centralized server. To enable the coexistence of multiple CR networks, spectrum broker will maintain information about all the spectrum resources.

1.3 Cognitive Cycle

A typical cognitive cycle of cognitive radio network is shown in the Figure 1.2 3, which includes identifying spectrum holes, selecting best available spectrum bands, coordinating spectrum access with other secondary users and vacating the channel when primary user appears 4.

Abbildung in dieser Leseprobe nicht enthalten

Figure 1.2: Cognitive Cycle of Cognitive Radio

Therefore the cognitive cycle of cognitive radio network consists of following four steps

1. Spectrum Sensing
2. Spectrum Decision
3. Spectrum Sharing
4. Spectrum Mobility

1.3.1 Spectrum Sensing

The main idea of cognitive radio is to identify the available spectrum through cognitive capability and re-configurability. The available spectrum is insufficient for the growing wireless users, so the basic challenge is to share the primary user’s spectrum with unlicensed or secondary users without causing any interference to the primary users which is shown in Figure 1.3 3. The unused spectrum by the primary users is called spectrum holes or spectrum white spaces 5. These spectrum holes can be utilized by the secondary users for transmitting their data. Whenever the primary user wants to utilize his frequency band then immediately the cognitive user has to move to another frequency band or it can stay in that band only by changing the transmission power and modulation technique to avoid interference.

In the process of sensing the spectrum, cognitive radio observes the surrounding environment to detect the unused licensed bands or spectrum holes and it also observes other cognitive users. This spectrum sensing can be done by single cognitive user or multiple cognitive users. Multiple cognitive user spectrum sensing improves the sensing performance.

1.3.2 Spectrum Decision

Spectrum decision is used to identify the available best spectrum bands but this decision is subjected to cause of any harmful interference to the primary users. Once the white spaces are identified, radio has to modify its operating frequency, modulation technique and the required technical parameters accordingly to transfer the data. Thereafter cognitive radio can send the information with good quality and without causing any harmful interference to the PU.

Abbildung in dieser Leseprobe nicht enthalten

1.3.3 Spectrum Sharing

Among all the parameters of cognitive radio, spectrum sharing holds an important aspect to consider. The available free spectrum can be utilized by the cognitive and non cognitive users through spectrum sharing. This spectrum sharing has to follow the policy rules of FCC i.e. secondary user can access the underutilized spectrum of the primary user without causing any interference and whenever primary user wants to utilize the same frequency spectrum, secondary user should vacate immediately.

1.3.4 Spectrum Mobility

Spectrum mobility is defined as the moment of secondary user from one spectrum hole to another. In order to avoid interference to the primary users, secondary user has to vacate the currently using radio frequency whenever the primary user starts operating in the same frequency band. Further the cognitive radio has to identify the alternate spectrum holes for data transmission.

1.4 Classification of Spectrum Sensing Techniques

The basic idea of the CR is to find out the available free spectrum through spectrum sensing. Therefore different spectrum sensing techniques are required to identify the available free spectrum in the radio environment with in less time and without causing interference to the PUs. The available spectrum sensing techniques are classified into three types 3. They are Transmitter Detection, Receiver Detection and Interference Temperature Management which are depicted in the Figure 1.4.

Abbildung in dieser Leseprobe nicht enthalten

Figure 1.4: Spectrum Sensing Techniques

1.4.1 Primary Transmitter Detection

Identifying the PUs transmitted signal at any given time is defined as the primary transmitter detection and it can be done by using hypothesis testing. In this technique, SU will detect the presence of the primary user with hypothesis 1 (Hi) and an absence where s(t) is the transmitted signal of a primary user, n(t) is the additive white Gaussian noise (AWGN), h(t) is the impulse response of the channel and y(t) is the secondary user’s signal at receiver. The available free spectrum is identified by detecting presence or absence of primary user on the channel. Hence to detect the primary user based on the hypothesis testing model, three transmitter detection techniques are used 7. They are Matched Filter Detection, Energy Detection and Cyclostationary Future Detection.

Abbildung in dieser Leseprobe nicht enthalten

Matched Filter Detection

In this method, initially secondary user will sense the channel and calculate the total signal strength of the same channel for a given signal of transmission 6. If the primary user is present then the signal received by the secondary user is the combination of primary user’s transmitted signal in addition with AWGN signal, otherwise it is only the AWGN signal. After identifying the received signal of a particular channel it is processed through a matched filter, where it is compared with the primary user’s transmitted signal which is already stored in it. The output of the matched filter is estimate with the decision threshold, which is shown in the Figure 1.5

Abbildung in dieser Leseprobe nicht enthalten

Figure 1.5: Block Diagram of Matched Filter Detection

It is observed that the primary user is absent when the matched filter output is less than the detection threshold and the channel can be utilized by the secondary user. whereas it is assumed that the primary user is present when the output of the matched filter is greater than the detection threshold and the channel cannot be utilized, so it has to sense the other channels. But here the secondary user needs an information of the PU like packet format, pulse shape, system order and modulation type. Hence it is very difficult to identify the available free spectrum if the secondary user does not have prior information about the primary user and is the major drawback in the matched filter detection method 2. The advantage of this method is that it will take less time to identify the presence or absence of the primary user.

Energy Detection

If the secondary user does not have prior information of the primary user, then matched filter method is not a preferred solution for identifying the primary user. Therefore energy detection method is considered as a suitable solution in which the primary user is detected by a secondary user based on the sensed energy. The block diagram of the Energy Detector is shown in the Figure 1.6.

Abbildung in dieser Leseprobe nicht enthalten

Figure 1.6: Block Diagram of Energy Detector

In this method, the received signal is transmitted through a band-pass filter with a bandwidth of W. It will allow signals of bandwidth W and limits the noise. The output of the band-pass filter is transmitted through a square law device which is used to improve the received signal strength. Each bit is integrated within the time interval of one bit by an integrator circuit and is compared with a predetermined threshold and the final decision i.e. presence or absence of the PU is given by the detection threshold. The threshold value will vary according to the channel conditions.

Some of the disadvantages of Energy Detection method are

1. Not able to differentiate signal and noise power because the primary user’s information is unknown to the secondary user.
2. Not useful to detect spread spectrum signals because of its wide bandwidth.
3. At low SNR this will take more time to achieve a given detection probability.

Cyclostationary Future Detection

This method exploits the periodicity in the primary user’s received signal in order to find the presence or absence of the primary user. The periodicity is generally embedded in sinusoidal carriers, hopping sequence or cyclic prefixes and pulse trains of the primary signals. The cyclostationary signals will contain the futures of periodical statistics and spectral correlation which are not found in the stationary noise and interference signals. This is the periodic nature of the primary users received signal 8.

Hence Cyclostationary Future Detection is a suitable solution for noisy and interference environments. Its performance is better than the Energy Detection method under low SNR regions, but it requires a prerequisite about the signal characteristics in order to differentiate the secondary user transactions from the various types of the primary user signals. It is used to identify the type of modulation used by the primary users.

Abbildung in dieser Leseprobe nicht enthalten

Figure 1.7: Block Diagram of Cyclostationary Future Detector

The block diagram of the Cyclostationary Future Detection is shown in the Figure 1.7. The received signal is processed through a band pass filter in order to limit the noise signal to band width of W. The output of band pass filter is split into two components, these two are multiplied with an exponential signal, obtained outputs are fed through a Fast Fourier Transform (FFT). Calculate the FFT for one signal and conjugate FFT for second signal. Compare these two signals to get correlated output. Average of correlated output signal is calculated using average over T. If the output of average over T is greater than the detection threshold, then PU is present. It has been used to differentiate signal power and noise power, and also it works better than the energy detection method. But it’s complexity is high and needs long processing time.

Limitations of Transmitter Detection

Shadowing and receiver uncertainty are the two major limitations in the transmitter detection 3. In this method, secondary users only have the information about the primary transmitter but not about the primary receiver. The transmitter detection faces the hidden node problem that limits its usage. Therefore the secondary user is unable to detect the presence of the PU when it is under shadowing region. On the other side, secondary user will detect the receiver by using a weak transmitted signal and it is unable to identify the presence of the PU, which is called as receiver uncertainty.

1.4.2 Receiver Detection

The drawbacks in primary transmitter detection can be eliminated by knowing the information about primary receiver. Hence to know the primary receiver’s information from its own architecture, the new spectrum sensing techniques are introduced. They are local oscillator leakage and sensor nodes for receiver detection.

[...]