Resource Allocation for a Massive MIMO Relay Aided Secure Communication

In this paper, we address the problem of joint power and time allocation for secure communications in a decode- and-forward massive multiple-input multiple-output (M-MIMO) relaying system in the presence of a passive eavesdropper. We apply the M-MIMO relaying technique to enhance the secrecy performance under very practical and adverse conditions, i.e., no availability of instantaneous eavesdropper channel state information (CSI) and only imperfect instantaneous legitimate CSI. We first provide a performance analysis of secrecy outage capacity, which reveals the minimum required number of relay antennas for achieving a positive secrecy outage capacity. Then, we propose an optimization framework to jointly optimize source transmit power, relay transmit power, and transmission time in each hop, with the goal of maximizing the secrecy outage capacity. Although the secrecy outage capacity is not a concave function with respect to the optimization variables, we show that it can be maximized by first maximizing over some of the variables, and then maximizing over the rest. To this end, we first derive a closed-form solution of optimal relay transmit power, afterward obtain that of optimal source transmit power, and then derive the optimal ratio of the first-hop duration to a complete transmission time. Moreover, several important system design insights are provided through asymptotic performance analysis. Finally, simulation results validate the effectiveness of the proposed joint resource allocation scheme.