Hybrid cooperative spectrum sensing scheme based on spatial-temporal correlation in cognitive radio enabled VANET

Abstract

Cognitive radio enabled vehicular ad-hoc networks (CR-VANETs) are one of the promising architectures in the future vehicular ad-hoc networks. In this study, the joint spatial-temporal correlation is exploited to improve decision accuracy of cooperative spectrum sensing (CSS) while reducing overhead introduced by cooperation in the CR-VANETs. Firstly, a theoretical method is presented to analyse the impact of spatial-temporal correlation on cooperative sensing performance when using soft combining in the CR-VANET. Then, the expression of the optimal probability of detection with respect to spatial- temporal correlation is given for a target probability of false alarm by employing likelihood ratio test. Additionally, the user selection problem is formulated as an efficient double-threshold optimisation problem by considering both sensing accuracy and stability to achieve the optimal probability of detection. Finally, a hybrid CSS scheme based on spatial-temporal correlation is designed for the CR-VANET. Simulation results reveal that the proposed scheme could achieve significant sensing performance gain by selecting a subset of secondary users for combination, and could reduce user selection frequency by employing spatial- temporal diversity.