Security in Cognitive Radio Networks (1402.1137v1)

Abstract: In this paper, we investigate the information-theoretic security by modeling a cognitive radio wiretap channel under quality-of-service (QoS) constraints and interference power limitations inflicted on primary users (PUs). We initially define four different transmission scenarios regarding channel sensing results and their correctness. We provide effective secure transmission rates at which a secondary eavesdropper is refrained from listening to a secondary transmitter (ST). Then, we construct a channel state transition diagram that characterizes this channel model. We obtain the effective secure capacity which describes the maximum constant buffer arrival rate under given QoS constraints. We find out the optimal transmission power policies that maximize the effective secure capacity, and then, we propose an algorithm that, in general, converges quickly to these optimal policy values. Finally, we show the performance levels and gains obtained under different channel conditions and scenarios. And, we emphasize, in particular, the significant effect of hidden-terminal problem on information-theoretic security in cognitive radios.