An Efficient Algorithm for a Weighted Cooperative Spectrum Sensing in Cognitive Networks

In this paper, a novel approach is proposed for a weighted cooperative spectrum sensing (WCSS) in cognitive radio networks (CRNs) aiming to maximize the probability of detection under a given probability of false alarm. In CRNs, Cooperative Spectrum Sensing (CSS) scheme is used to get over the problem of hidden terminal, fading and shadowing. The proposed algorithm can be applied for single and double thresholds energy detector. Our goal is to have an efficient WCSS with less complexity and high performance. The existing works showed that finding the optimal weights for probability of detection maximization is a difficult problem. Therefore, we propose a closed form suboptimal solution using the generalized eigenvalue that outperform some of the existing works in terms of performance and complexity. The proposed approach is compared to particle swarm optimization (PSO), equal gain combining (EGC), modified deflection coefficient (MDC) in terms of complexity and performance. Our results show that the proposed approach outperforms all these methods especially when the users receive different signal to noise ratio (SNR).