Robust node selection for cooperative spectrum sensing with malicious users

This paper proposes a practical sensing node selection for cooperative spectrum sensing in a cognitive radio network, where malicious users could falsify sensing data. Previously reported sensing node selection techniques are applicable to scenarios where sensing nodes are assumed to send reliable reports, and do not have the mechanisms to check their trustworthiness. In this paper we propose a robust sensing node selection algorithm which evaluates the level of trustworthiness via a weighted consistency check with reports of trusted nodes. Simulation results show that existing node selection techniques would include malicious users in cooperative sensing such that the performance is significantly degraded. Our proposed algorithm is shown to exhibit robustness and effectiveness in the presence of malicious users, while preserving efficiency in friendly environments.