A cooperative spectrum sensing scheme based on the Bayesian reputation model in cognitive radio networks

Abstract

Cooperative spectrum sensing is regarded as an efficient way to detect the spare spectrum in cognitive radio networks (CRNs), where malicious secondary users (SUs) can launch the spectrum sensing data falsification (SSDF) attack to interfere with the detection and threat the CRN. In order to defend against the SSDF attack, it is necessary for the fusion center to distinguish the trustworthiness of the reports from SUs. In this paper, we propose a cooperative spectrum sensing scheme based on the Bayesian reputation model for CRNs. The key idea is that the cooperation is viewed as a service-evaluation process, and SUs' reputation degrees reflect their service quality. In our proposed scheme, reputation degrees of SUs are updated based on the Bayesian reputation model, and the behaviors of malicious SUs can be tracked effectively. Simulation results show that our proposed scheme defends against various types of SSDF attack, especially the probability SSDF attack.