Securing Cognitive Radio Networks with Distributed Trust Management against Belief Manipulation Attacks

Abstract

In Cognitive Radio Networks (CRNs), Cognitive radios (CRs) learn from their environment and adapt to the environment based on their learned beliefs accordingly. Malicious nodes may exploit the cognitive engine of CRs, and conduct belief manipulation attacks to degrade the network performance. In this paper, we address the problem of belief manipulation attacks and develop a distributed trust management strategy to detect and mitigate such attacks in CRNs. Specifically, we first study the impact of malicious behaviors to the network performance, and define trust evaluation metrics to capture malicious behaviors. We then illustrate how to incorporate distributed trust management to mitigate the effectiveness of belief manipulation attacks to enhance the security in CRNs. Performance evaluation results show that the network end-to-end throughput is significantly improved compared to the case when all users are by default trusted to be normal users.