MIMO Secure Communication with Reconfigurable Intelligent Surface: Finite-Alphabet Inputs

The existing works in the field of reconfigurable intelligent surface (RIS)-aided physical-layer security considered the Gaussian input, which is however infeasible in practical systems. This paper investigates the secrecy rate (SR) maximization problem in RIS-aided multiple-input multiple-output wireless transmission with finite-alphabet inputs. Specifically, a closed-form lower bound of the SR is firstly derived and is used as the objective function, so as to deal with the exact but intractable expression of the SR. Then, the transmit precoder and RIS's phase shifts are alternately optimized by using the Lagrange multiplier method and the Hooke-Jeeves algorithm, respectively. Numerical results show that the RIS can significantly improve the SR for finite-alphabet-input transmission, even if RIS's phase shifts are subject to the finite-resolution constraint.