Random matrix cooperative spectrum sensing for clustered sensors using Neyman-Pearson Fusion

In this paper we use a new approach to applying the random matrix properties of cognitive radio to spectrum sensing in cognitive radio for clustered sensors, where the Secondary User (SU) sensors within a cluster are assumed to be experiencing the same noise variance and the same Primary User (PU) Signal to Noise Ratio (SNR). Pointing out some recent works on the application of Random Matrix Theory (RMT) in spectrum sensing, we suggest slight but effective changes to the previously mentioned detection strategies, which enables us to examine the idea more comprehensively from a detection theory point of view. We apply the proposed detection strategy as our spectrum sensing scheme within clusters, we then assume to have a Neyman Pearson Fusion Center where the cluster decisions are combined to obtain the final decision as our spectrum sensing. Simulation results show that with no prior knowledge about the PU signal or the noise distribution, our proposed scheme performs quite desirably.