Power Allocation Strategies for Secure Spatial Modulation

In this paper, power allocation (PA) strategies in secure spatial modulation networks, are investigated under the total power constraint. Considering that there is no closed-form expression for secrecy rate (SR), an approximate closed-form expression of SR is derived as an efficient metric to optimize PA factor, which can greatly reduce the computation complexity. Based on this expression, a convex optimization (CO) method of maximizing SR (Max-SR) is proposed accordingly. Furthermore, a method of maximizing the product of signal-to-leakage and noise ratio (SLNR) and artificial noise-to-leakage and noise ratio (Max-P-SAN) is proposed to provide an analytic solution for PA factor with extremely low complexity. Simulation results demonstrate that the SR performance of the proposed CO method is close to that of the optimal PA strategy with exhaustive search, and is better than that of Max-P-SAN in the high signal-to-noise ratio (SNR) region. However, in the low and medium SNR regions, the proposed Max-P-SAN slightly outperforms the proposed CO scheme in terms of SR performance.