STAR-RIS Aided Secure Communication for Endogenous Wiretap Broadcast Channels With Two Users

• Published in: IEEE transactions on vehicular technology Vol. 74; no. 3; pp. 4221 - 4234
• Main Authors: Shang, Liming, Jiang, Weiheng, Nie, Jiangtian, Xiong, Zehui, Chen, Chen
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.03.2025; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Array signal processing; Beamforming; Communication; Communications systems; confidential messages; Convexity; Downlink; Eavesdropping; Electronic surveillance; Forensics; Optimization; Oral communication; Physical layer; physical layer security; Protocol; Protocols; Reconfigurable intelligent surfaces; Splitting; STAR-RIS; Surveillance; Transmission; Wireless networks; Wiretapping
• ISSN: 0018-9545; 1939-9359
• DOI: 10.1109/TVT.2024.3486356

In this paper, we study a simultaneously transmitting and reflecting reconfigurable intelligent surface (STAR-RIS) aided two-user endogenous wiretap communication system, in which a base station (BS) has confidential communication requirements with two users located in the transmitting and reflecting zone of STAR-RIS, respectively. Specifically, these two users are both legitimate users and mutual eavesdroppers. In this system, we investigate the achievable rate for these two users under three typical STAR-RIS protocols, i.e., energy splitting (ES), time splitting (TS) and mode switching (MS). Based on that, a series of weighted sum secrecy rate (WSSR) maximization problems subject to the constraints of BS transmit power and the transmitting and reflecting coefficients (TRCs) of STAR-RIS are formulated, which consider to jointly optimize the active and passive beamforming vectors at the BS and STAR-RIS, respectively. Due to the non-convexity of these problems and the coupled decision variables, the path-following method is applied to approximate the non-convex objective function and then, the alternating optimization (AO) method is utilized to decouple the decision variables. After that, relaxation and penalty convex concave procedure (PCCP) are adopted to solve the decomposed subproblems. In further, we analyze the computational complexity and convergence of the proposed AO-based algorithms. In the end, some numerical simulations have been done to demonstrate the performance of the proposed algorithms, and the results show that the ES protocol always obtains the best performances, followed by the MS protocol, and the TS protocol obtains the worst performance.