Reconfigurable Intelligent Surface Aided Physical-Layer Security Enhancement in Integrated Sensing and Communication Systems

Integrated sensing and communication (ISAC) has the problem of information leakage when using the ISAC waveforms to sense the surroundings, thereby posing a risk for potential eavesdropping. To tackle this issue, this paper proposes a novel reconfigurable intelligent surface (RIS) aided beamforming design for physical-layer security enhancement in the ISAC system. Considering that an eavesdropper (EVE) possibly appears in a certain region and the wireless channel associated with the EVE is unknown, an approximate ergodic achievable secrecy rate (EASR) is derived mathematically. Then, an optimization problem is formulated to maximize the approximate EASR, while ensuring the minimum communication performance requirement for the legitimate user and the minimum sensing performance requirement for the target. To solve the non-convex problem, a novel optimization approach is developed to alternatively obtain the solutions of the transmit beamformer, the covariance matrix of the artificial noise (AN), the RIS phase-shift matrix, and the receive beamformer. Specifically, the first two are optimized by constructing a lower-bound surrogate function of the original objective. The RIS phase-shift matrix is optimized by designing a successive convex approximation (SCA) based approach. The receive beamformer is optimized by solving a Rayleigh-quotient subproblem. Finally, the overall optimization algorithm is designed and theoretically proved to be convergent. Simulation results validate the tightness of the approximate EASR, and show that our proposed design outperforms the existing benchmarks in terms of the security or sensing performance.