DCA approaches for simultaneous wireless information power transfer in MISO secrecy channel

This paper addresses the emerging practical requirement that wireless information and power transfer are employed simultaneously in a multiple-input single-output (MISO) secrecy channel. Transmit beamforming without artificial noise and that with artificial noise are considered. In addition, perfect and imperfect channel state information of both the legitimate receivers and eavesdroppers are investigated. These four scenarios correspond to four optimization problems. The considered problems aim to maximize the secrecy rate with constraints on the transmit power and the harvested energy at each receiver. Since these optimization problems are nonconvex, we propose the efficient algorithms based on Difference of Convex (DC) functions programming and DC Algorithm (DCA )-an innovative approach in nonconvex optimization. Exploiting the special structure of these problems, we utilize the semidefinite relaxation technique and S-procedure to reformulate them as general DC programs. The efficient algorithms based on general DCA are developed to solve the corresponding general DC programs. Numerical results are presented which illustrate the effectiveness of our proposed algorithms, they outperform the existing approaches in terms of both quality and rapidity.