Performance Evaluation of Local and Cooperative Spectrum Sensing in Cognitive

Radio

by

Harish Barvekar

Roll No-11EC64R17

(M.Tech TSE)

Under the supervision of

Dr. Saswat Chakrabarti

Department of Electronics & Electrical Comm. Engg.

Indian Institute of Technology, Kharagpur, India

2 Contents

Introduction

Problem Definition

Implementation of Energy detection technique

Implementation of One-Order Cyclostationary Feature Detector

Implementation of Two-Order Cyclostationary Feature Detector

Implementation of Cooperative spectrum sensing using energy detector

Implementation of spectrum management and sharing concept

Conclusion

Future Work

References

3 Background

most of the radio frequency spectrum inefficiently utilized

spectrum utilization depends strongly on time and place

fixed spectrum allocation wastes resources

improved efficiency by allowing unlicensed users to exploit spectrum whenever it would not cause interference to licensed users

4White Spaces

McHenry. M. A., “NSF Spectrum occupancy measurements project summary,” Shared Spectrum Company, Tech. Rep., Aug.

2005.

5 Functions of Cognitive Radio

Spectrum Sensing

Spectrum Management

Spectrum Sharing

Spectrum Mobility

6 Types of spectrum sensing

7 Problem Definition

Implementation and performance evaluation of local spectrum sensing techniques like Energy detector, One-order Cyclostationary feature detection and Two-order Cyclostationary feature detection

Implementation of Cooperative spectrum sensing through energy detector

Implementation of spectrum management and spectrum sharing characteristics with the help of posix Pthread program in Linux platform

8

In the spectrum sensing process, the sensing users observe signals under the following two hypotheses:

: absence of primary user at the spectrum band of interest,

: presence of primary user at the spectrum band of interest

r(t)= n(t) (white space)

r(t) = s(t) + n(t) (occupied)

where,

r(t) - received signal at the cognitive radio

s(t) – transmitted signal from the primary user

n(t) – AWGN noise

Signal Model

9

Received Signal Strength [6]

10Performance of a spectrum sensing technique is generally measured in terms of the probability of detection(Pd) probability of false alarm(Pf) and probability of miss detection(Pm) which are define as

Pd= Pr (/)

Pf = Pr (/)

Pm= Pr (/)

The probability of detection can be written in terms of probability of miss detection as

Pd= 1- Pm

11

Implementation of Energy detection technique

Block diagram of an energy detector

𝑌 { ¿ 𝜒2𝑢2 𝐻0

¿ 𝜒 2𝑢2 (2𝛾 ) 𝐻1

Following the work of Urkowitz [13], Y may be shown to have the following distribution

whereand denote the central and non-central chi-square distributions, respectively, each with 2u degrees of freedom and a non-centrality parameter of is the SNR.

12

the PDF of Y under two hypotheses may be written as [7]

where is the gamma function and (.) is the v-th order modified Bessel function of the first kind

The detection and false alarm probabilities are computed by

= (Y>

= (Y>)

Solving we get,

whereis the incomplete gamma function and represent the Marcum Q-function

13

1. Initialize number of samples N, prob. of false alarm and avg. S/N

2. Initialize counter to zero

7. Counter = counter + 1

3. Generate an input signal and add additive white Gaussian noise to it

4. Calculate the threshold value using known set of probability of false alarm

6.Is Energy of s/g >threshold

8.Calculate the probability of detecting the signal by dividing the final value counter with the number of

Monte –Carlo iterations

9.Calculate the value probability of miss detection by subtracting the values of probability of detection from

the integer 1

10.Plot the graph between probability of false alarm and probabilityof miss detection

5. Calculate the energy of the noisy input signal

Repeat Steps 3-6 for Monte-

Carlo times

Flowchart of algorithm

14

Simulation Result

Complementary Receiver Operating Characteristic (CROC) curve of Energy detection technique over AWGN channel (a) theoretical (b) as obtained through simulation

15

Implementation of One-Order Cyclostationary Feature Detector

Consider a sine signal s(t) such that

s(t) = Asin(2)

where A is the envelope, is the periodic frequency and is the phase of signal

In the transmission of s(t) through an AWGN channel,

x(t) = s(t) + n(t)

The mean function of x(t) can be written as

= E[x(t)] =s(t)

One-order Cyclostationary feature detection

16For any instant of time t and any value of integer P, we can calculate mean for test statistics

Y =

The probability of false alarm and detection is given by [09]

OFD= exp

where is the generalized Marcum Q-function and is the instantaneous signal to noise ratio and is the variance with the non-centrality parameter A.

17

Simulation Result

Complementary Receiver Operating Characteristic (CROC) curve of one-order Cyclostationary detection technique over AWGN channel (a)

theoretical (b) as obtained through simulation

18

Implementation of Two-Order Cyclostationary Feature Detector

The input signal is time shifted with a period . The time shifted signal is auto-correlated using the equation

(t) = *

Two-order Cyclostationary detection technique

19 The threshold value for two order cyclostationary detection can be obtained [09] from the equation

= exp

where N is the number of samples, is the threshold and is the variance of the input signal

Probability of detection is given by [09]

=

whereis the instantaneous signal to noise ratio of the received signal. And the value of can be calculated from the equation

=

20

Simulation Result

Complementary Receiver Operating Characteristic (CROC) curve of two-order Cyclostationary detection technique over AWGN channel

(a) theoretical (b) as obtained through simulation

21

Implementation of Cooperative spectrum sensing using energy detector

Every CR performs its own local spectrum sensing measurements independently and then makes decision on whether the PU is present or not

All of the CRs forward their decisions to a common receiver

The common receiver fuses the CR decisions and makes a final decision to infer the absence or presence of the PU

22

Each CR calculates its individual probability of miss detection using energy detection technique

The fusion center receives n decisions from cognitive users and adds up all the decisions from all the CRs. OR rule is used for decision making.

Detection probability and false alarm probability represented by and respectively for cooperative sensing can be given as [10]

where n is the number of CR used. and are given by the equations of energy detector stated in above slides

23

Simulation Result

Receiver Operating Characteristics (ROC) curve of Cooperative spectrum sensing with OR rule under AWGN channel

5 CRs are used for local sensing

24

Receiver Operating Characteristics (ROC) curve of Cooperative spectrum sensing with OR rule under AWGN Channel with CR=1, 2, 3,4 and 5

25

Implementation of spectrum management and sharing concept

Challenges after detection of spectrum holes, how to allocate channels to the secondary

users

how to handle the situation, when primary user comes back for transmission

Selection criteria User is selected by their priority

if priority equals, then user is selected by their burst time

26

Flowchart of Spectrum sharing program

S1B3

S3B6

S2B5

S1B4

Communication channel

Priority queue

Primary user

Secondary user

S3B6

S2B5

S4B9S1B3

Communication channel

Priority queue

Primary user

Secondary user

S3B6

S2B5

S4B9

S1B3

S5B7

Communication channel

Priority queue

Primary user

Secondary user

S3B6

S2B5

S4B9

S1B3

S5B7

Communication channel

Priority queue

Primary user

Secondary user

S3B6

S2B5

S4B9

S5B7

S1B2

Communication channel

Priority queue

Primary user

Secondary user

S3B6

S2B5

S4B9

S5B7

S1B2

Communication channel

Priority queue

Primary user

Secondary user

S3B6

S2B5

S4B9

S5B7

S1B2

P1B6

Communication channel

Priority queue

Primary user

Secondary user

S3B6

S2B5

S4B9

S5B7

S1B2

P1B6

Communication channel

Priority queue

Primary user

Secondary user

S3B6

S2B5

S4B9

S5B7

S1B2

P1B5P1B6

Communication channel

Priority queue

Primary user

Secondary user

S3B6

S2B5

S4B9

S5B7

S1B2

P1B5

Communication channel

Priority queue

Primary user

Secondary user

Simulation Result

40

Conclusion

The complementary receiver operating characteristic curve for the spectrum sensing techniques are studied

The receiver operating characteristic curve for cooperative spectrum sensing using energy detector is also studied

The concept of spectrum management and sharing is studied by a software Pthread program

41

Future Work

The work can be extended to a scenario where multiple primary users are involved

Security can be enhanced by detecting the malicious user in the Cognitive radio network and remove their decision

Concept of primary user emulation attack can also be introduced in cooperative spectrum sensing

42

References

[1] J. Mitola and G. Q. Maguire, “Cognitive radio: Making Software Radios More Personal,” IEEE Pers, Commun., vol. 6, pp. 13–18, Aug. 1999.

[2] McHenry. M. A., “NSF Spectrum occupancy measurements project summary,” Shared Spectrum Company, Tech. Rep., Aug. 2005.

[3] I. F. Akyildiz, Y. Altunbasak, F. Fekri, and R. Sivakumar, “AdaptNet: an adaptive protocol suite for the next- generation wireless internet,” IEEE Commun. Mag., pp. 128–136, Mar. 2004.

[4] Federal Communications Commission, “Notice of Proposed Rule Making and Order,” Rep. ET Docket no.03-322, Dec. 2003.

[5] S. Haykin, “Cognitive radio: Brain-empowered wireless communications,” IEEE J. Sel. Areas Commun., vol. 23, no. 2, pp. 201 - 220, Feb. 2005

[6] M. Ghozzi, M. Dohler, F. Marx, and J. Palicot, “Cognitive radio: methods for the detection of free band,” Comptes Rendus Physique, Elsevier, vol. 7, pp. 794–804, Sep. 2006.

[7] Fadel F. Digham, Mohamed-Slim Alouini, and Marvin K. Simon,” On the Energy Detection of Unknown Signals over Fading Channels” IEEE Transactions On Wireless Communications, vol. 7, no. 12, pp. 3575 – 3579, Dec. 2008

[8] EkramHossain, DusitNiyato and Zhu Han, “Dynamic Spectrum Access and Management in Cognitive Radio Networks”, Cambridge university Press, 2009

[9] Y. Wen-jing, Z. Bao-yu, M. Qing-min,” Cyclostationary property based Spectrum sensing algorithms for primary detection in Cognitive Radio systems,” Institute of Signal Processing and Transmission, Nanjing 2100003, China, 2008

43

[10] C. Sun, W. Zhang, and K. B. Letaief, “Cluster-Based Cooperative Spectrum Sensing in Cognitive Radio Systems,” Hong Kong University of Science and Technology, 2007

[11] Lei Zhang and Zhijun Xiao, “Performance Analysis of Cooperative Spectrum Sensing Algorithm for Cognitive Radio Networks,” International Conference on Computer Design and Applications (ICCDA 2010), vol.4, pp.V4-557- V4-560, 25-27, June 2010

[12] Xuping. Zhai, Jiango. Pan, “Energy-Detection Based Spectrum Sensing for Cognitive Radio, Wireless, Mobile and Sensor Networks,” IET Conference, 2007

[13] H. Urkowitz, “Energy Detection of Unknown Deterministic Signals,” IEEE, vol.55, pp.1606, April 2002

[14] Chao Chen, Hongbing Cheng and Yu-Dong Yao, “Cooperative Spectrum Sensing in Cognitive Radio Networks in the Presence of the Primary User Emulation Attack,” IEEE Transactions on Wireless Communications, vol.10, Issue 7,  pp.2135 – 2141,July 2011

[15] F. Visser, G. Janssen, P. Pawelczak, “Multinode Spectrum Sensing Based on Energy Detection for dynamic Spectrum Access,” IEEE, pp. 1394-1398, 2008

[16] D. Duan, L. Yang and J. C. Principe, “Cooperative Diversity of Spectrum Sensing for Cognitive Radio Systems,” IEEE transactions on signal processing, vol. 58, no. 6, June 2010

[17] Y. Gao and Y. Jiang, “Performance Analysis of A Cognitive Radio Network With Imperfect Spectrum Sensing”, IEEE Infocom 2010, pp.1-6, Issue date: 15-19, March 2010

[18] Paisana F., Prasad N., Rodrigues A., Prasad R., “An alternative implementation of a cyclostationary detector,” Wireless Personal Multimedia Communications (WPMC), pp. 45-49, 2012

Thank You