Precoding and Artificial Noise Design for Cognitive MIMOME Wiretap Channels

In this paper, we study the secrecy precoding problem for a cognitive multiple-input-multiple-output multiple-eavesdropper (MIMOME) wiretap system. The problem is studied with an artificial noise (AN)-aided precoding scheme. First, we place AN in the null space of the legitimate receiver's channel matrix and formulate a secrecy rate maximization (SRM) problem, which is subject to both an interference power constraint imposed to protect the primary user (PU) and a maximum transmit power constraint available for the secondary transmitter. Second, we drop the null space constraint and reformulate the SRM problem. Because the formulated SRM problems naturally constitute difference-convex-type programming problems, we solve them by employing a successive convex approximation method, where the nonconvex parts of each problem are approximated by their first-order Taylor expansion. Thus, the SRM problems can be iteratively solved through successive convex programming of their convexified versions. Results show that our algorithms can achieve a satisfactory solution with guaranteed convergence.