Optimal cooperative spectrum sensing over composite fading channels

Abstract

In this paper, we study cooperative spectrum sensing over composite fading channels. First, we consider the Mixture of Gaussian distribution to model the composite channel statistics and derive a simple generic approximation for the average probability of detection which can be efficiently applied to any composite fading channel. Second, we derive the optimal voting rule for hard combining in cooperative spectrum sensing over composite fading channels. In particular, we derive an exact closed form expression for the optimal decision fusion rule (k-out-of-N) that minimizes the total sensing error in cognitive radio networks (CRNs). This expression enjoys both simplicity and generality, making it applicable to any CRN, regardless of the channel considered. Numerical results validate the proposed analysis and show that the proposed combining rule efficiently minimizes the CRN's total error and significantly outperforms the AND, OR and majority rules that are usually considered.