Performance of Cooperative Spectrum Sensing with Soft Data Fusion Schemes in Fading Channels

Cognitive radio (CR) systems need to detect the presence of licensed primary user (PU) reliably. Cooperation among multiple CRs helps to enhance the reliability of detection of the PU in case of unreliable decision by a single CR due to channel uncertainties. In this paper, we consider cooperative spectrum sensing (CSS) based on energy detection in cognitive radio networks (CRN) which uses soft combination of the observed energy values from different CRs. More precisely, we study the performance of CSS with several soft data fusion schemes namely, (a) Square law selection (SLS) (b) Square law combining (SLC) (c) Maximal ratio combining (MRC) that can be implemented at fusion center (FC). The performance of CSS has been assessed under several cases of sensing (S) channels such as AWGN, log-normal shadowing, Rayleigh and Rician fading channels. Comparative performance of CSS for various soft data fusion schemes under different fading channels has been studied for various values of average S-channel SNRs, time-bandwidth products, different number of CRs. The impact of shadowing and fading parameters on missed detection performance of CSS is indicated. Further, the performance comparison between soft data fusion schemes and hard decision fusion schemes is also highlighted.