Secrecy rate maximization in a cognitive radio network with artificial noise aided for MISO multi-eves

Abstract

In this paper, we consider beamforming design for an underlay cognitive radio multiple-input single-output broadcast channel, where a pair of secondary users coexists with multiple primary receivers. There exist multiple malicious eavesdroppers who attempt to overhear the confidential messages from the secondary system. When the channel state information of the eavesdroppers can only be obtained in the statistical sense, we transform the constraint which results from the statistical information of the passive eavesdroppers into a linear matrix inequality and convex constraint. To improve the secrecy rate of the secondary system, we aim to design a jamming noise to degrade the eavesdroppers. The main objective of the design is to maximize the secrecy rate of the secondary system while satisfying all the interference power constraints at the primary users and per-antenna power constraint at the secondary transmitter. The original problem is a nonconvex program, which can be reformulated to a convex program by applying the rank relaxation method. To this end, we prove that the rank relaxation is tight and it can be efficiently solved. Moreover, to develop an efficient resource allocation scheme we transform the relaxed problem into an equivalent problem based on a duality result.