Secure Symbol-Level Precoding Designs via Intelligent Reflecting Surface

Abstract

Intelligent reflecting surface (IRS) as a stand-out technology that can intelligently adjust the wireless propagation environment. In addition, the symbol-level precoding (SLP) technology proposed in recent years uses known interference to enhance the quality of legitimate communication and improve physical layer security (PLS) symbol by symbol designed. This paper investigates the IRS- assisted secure MU-MISO communication system. Aiming at minimizing the transmission power, the transmit SLP matrix at the base station (BS) and the phase shifts matrix at the IRS are jointly designed under the constraints of legitimate user constructive interference (CI), eavesdropper destructive interference (DI) and IRS unit-norm. A low complexity gradient descent (GD) algorithm and manifold based Riemann conjugate gradient (RCG) algorithm are proposed to deal with the SLP problem and IRS reflection design problem respectively. Numerical results illustrate that the system performance is significantly advanced by deploying IRS in the MU-MISO secure communication system and using SLP technology.