Abstract: Cooperative spectrum sensing (CSS) improves the accuracy of spectrum detection. However, when there are a large number of malicious users in secondary users, malicious users will launch large-scale attacks, which will seriously affect the CSS results. A user selection and belief propagation based dual defense scheme for large-scale intrusion is proposed. Based on the results of local spectrum sensing, a user selection scheme is used to pick out reliable secondary users which include honest users and malicious users pretending to be honest. Then, the belief values of secondary users will be calculated by the belief propagation scheme and the mean of belief values is calculated with the reliabilities as the weights and compared with the preset threshold to further detect the malicious users. Simulation results show that the proposed scheme has better detection and convergence performance than the existing detection algorithm such as simple reputation-based scheme, mean-based scheme and lightweight cloned-node detection algorithm.

Key words: cooperative spectrum sensing （CSS）, intrusion prevention, user selection, belief propagation, reputation