Abstract
—Physical layer security (PLS) is a promising technique to improve the security of wireless communications. However , when the legitimate user's and eavesdropper's channels are strongly correlated, directly applying PLS might no longer be a valid approach. In this paper, by introducing the reconfigurable intelligent surface (RIS), we study how to design access point (AP) and RIS beamforming and deploy RIS to improve the security under strong channel correlation. Furthermore, by taking into consideration the " multiplicative fading " effect, we formulate the secrecy rate (SR) maximization problem in the passive and active RIS cases. Next, we propose a semidefinite program (SDR)-based alternative optimization (AO) algorithm for each case, respectively. However, the computational complexity of the SDR approach is prohibitive for the large-size RIS. To tackle this issue, we respectively develop the low-complexity minorization maximization-based and primal-dual subgradient-based AO algorithms for two cases. Finally, we analyze the effect of the RIS deployment on the SR, and simulations results demonstrate the effectiveness of the proposed schemes.