Secure green communication for amplify-and-forward relaying with eavesdroppers

In this paper, the secure green communication over amplify-and-forward (AF) relaying channels is investigated by using the metric of secure energy efficiency (EE), defined as the number of bits securely delivered per unit energy consumption. We maximize the secure EE with the given maximum power and the minimum secrecy rate requirements. The difficulty of the problem comes from the nonconvexity of both the objective function and the secrecy rate constraint. Therefore, a suboptimal solution scheme, based on fractional programming, dual decomposition, and DC (difference of convex functions) programming, is addressed to solve this problem iteratively. The proposed solution scheme composed of three-layer iterations transforms the primal problem into easier subproblems at each iterative layer, such that the resulting subproblems can be solved by the corresponding optimization methods mentioned above. The simulation results demonstrate that the secure EE of the proposed scheme is superior than that obtained by the conventional schemes with a small decrease of the average secrecy rate.