Preventing Collusion Attacks in Cooperative Spectrum Sensing

Cooperative spectrum sensing is prone to attacks by malicious users. These attacks become more severe when launched in collusion. In this paper, a collision penalty based attack prevention method is proposed to defend individual and collusion attacks in all the possible attack scenarios. We identify the possible attack strategies in each scenario and prevent them with appropriate bounds on the collision penalty. Simulation results verify the effectiveness of the proposed scheme.