Cooperative Spectrum Sensing Over Hyper-Rayleigh Fading Channels in Multiple Antenna Based Cognitive Radio Networks

Abstract

Wireless communication systems have opened new dimensions in communications. People can be reached at any time and at any place. In today's world one of the most widely used forms of communication is wireless communication. But the demand of wireless applications are increasing and the number of unlicensed users are increasing which makes the problem of development for the wireless communication and the problem is known as spectrum scarcity. So to overcome the problem of spectrum scarcity, A tool known as the cognitive radio has been proposed. Primary users will not use the spectrum all the time. So to make the efficient utilization of the spectrum, cognitive radio helps the Secondary users which are known as the licensed users or the cognitive users to make use of the licensed spectrum band when the primary user is not using the licensed spectrum band. Cognitive radio (CR) is a very intelligent tool which sense the spectrum very intelligently and helps in making the efficient utilization of the available spectrum band. Spectrum sensing is the important function of CR system and is very crucial. The various techniques of spectrum sensing has been studied and the various detectors has been used. Energy detector, matched filter, cyclostationary detectors has been studied. Energy detector is easy to implement since it is less complex and cyclostationary detector is complex to implement but the accuracy of cyclostationary detector is much better than the energy detector. In this thesis, performance analysis of cooperative spectrum sensing employing improved energy detector over TWDP channel is evaluated and compared with threshold optimization and CFAR methods. In cooperative sensing, at each cognitive radio multiple receiving antennas has been used. Detection performance of the proposed schemes is analyzed in terms of error detection probability w.r.t. SNR. Simulation results shows that by using cooperative sensing, error detection probability can be reduced or in other words SNR gains of 14 dB and 3 dB can be achieved in comparison to CFAR and optimum threshold method respectively.