Optimization of Pre-Processing Filter for Time-Reversal Multi-User Secure Transmission Systems Based on Artificial Noise

Abstract Physical layer security is a way to realize the secure transmission of information by employing the characteristics of wireless channels. Thanks to its spatial and temporal focusing property, the time-reversal (TR) transmission can achieve effective secure communications even when the transmitter is only equipped with one antenna. This paper studies the design and optimization of the security scheme based on artificial noise (AN) with or without the eavesdropper's channel state information (ECSI) in a TR multi-user downlink multiple access system. The null-space AN is adopted when ECSI is unavailable, and the corresponding optimization problem is to minimize the signal power under each user's signal-to-interference-plus-noise ratio (SINR) target constraint, so that the power used to send AN is maximized and the interception rate of the eavesdropper is minimized. The problem is solved by transforming it into a semidefinite program (SDP) with semidefinite relaxation (SDR). The pre-processing filter and AN are jointly optimized to maximize the sum secrecy rate when ECSI is available. The problem is solved by transforming it into a convex program by using SDR and the first-order Taylor approximation. Simulation results show that the sum secrecy rate of the proposed scheme is significantly higher than that of the conventional TR transmission system with or without AN.