Robust Beamforming Design for Achievable Secrecy Rate Maximization in IRS-Aided Satellite Systems

Abstract

This paper investigates a robust secure transmission design for enhancing the physical layer security (PLS) in an intelligent reflecting surface (IRS)-aided satellite system. Especially since it is challenging to obtain the perfect channel state information (CSI) related to the IRS links, we consider here two different scenarios. Specifically, (1) perfect legitimate users' CSI (LCSI) and imperfect eavesdroppers' CSI (ECSI) are adopted. (2) the case of imperfect LCSI and imperfect ECSI. Under both scenarios, we formulate an optimization problem aiming to maximize the minimal achievable secrecy rate (ASR) among legitimate users subject to quality-of-service requirements and satellite transmit power budget. In order to tackle this nonconvex optimization problem, we propose a robust beamforming (BF) algorithm combining second-order Taylor series expansion and Bernstein-type inequality to jointly optimize transmit power of satellite and phase shifts of IRS. Finally, numerical results confirm the robustness and superiority of the proposed scheme under different CSI uncertainty scenarios.