Joint Beamforming and RIS Design Against a Multi-Antenna Active Eavesdropper

Abstract

This paper investigates a secure integrated sensing and communication (ISAC) system assisted by a reconfigurable intelligent surface (RIS). We consider a challenging threat scenario involving a multi-antenna full-duplex eavesdropper that not only wiretaps confidential communication signals but also actively emits jamming signals toward both legitimate communication users and the sensing target. To safeguard the system against such simultaneous eavesdropping and jamming attacks, artificial noise (AN) is embedded into the base station (BS) transmit signals. We formulate a joint optimization problem for the transmit design and RIS phase shifts, aiming to maximize the sum secrecy rate (SSR), subject to the minimum signal-to-interference-plusnoise ratio (SINR) requirements of the communication users and a minimum echo SINR required for reliable target sensing. To tackle this highly coupled and intrinsically non-convex problem, we employ a low-complexity alternating optimization (AO) framework to obtain an effective suboptimal solution. Simulation results unveil that the proposed design can achieve a favorable trade-off between communication performance and sensing quality, and further reveal that the exploitation of AN and RIS offers great potential for enhancing the overall security performance of ISAC systems.