Optimal Fusion Techniques for Cooperative Spectrum Sensing in Cognitive Radio Networks

Cooperative Spectrum Sensing (CSS) is a critical component in cognitive radio networks (CRN). It is a scheme used to opportunistically scan the spectrum by multiple CR devices to make collaborative decisions on the presence or absence of the Primary User (PU). It improves the performance on spectrum utilization by exploring the spatial diversity of the cognitive radio networks. Both the selection criterion of cooperating secondary users (SUs) and the fusion schemes used for CSS affect the detection probabilities of the PU. This paper investigates the performance of AND, majority and OR rule fusion techniques. The performance based on an optimal number of collaborating SUs in majority fusion rule is considered and comparison made against OR and AND rules. The optimal number of SUs collaborating in sensing is investigated using Newton Raphson Method. The performance results indicate that majority fusion rule has better detection capability, low total error probability as compared to the OR and AND rules at low SNR.