Secure beamforming for cognitive radio networks with artificial noise

In this paper, we address the secrecy problem for a cognitive radio network (CRN) by using physical layer security (PHY-security) technology. In order to effectively enhance wireless security, we propose to adopt artificial noise (AN) to impair the quality of the interception signal. For such a secure CRN, it is a challenging issue to design the beams for both the desired signal and the AN, so as to maximize the secrecy rate while satisfying the interference constraint to primary user (PU). Since the formulated problem is nonconvex, we adopt an alternative optimization method combined with one-dimensional linear searching to design optimal secure beams. Furthermore, we propose a low-complexity secure beamforming scheme by confining the AN in the orthogonal space of the legitimate channel. It is found that the low-complex scheme can asymptotically approach the same performance as the optimal one, when the number of antennas is large. Finally, simulation results validate the effectiveness of the two proposed schemes.