Robust Secure Hybrid Beamforming for Active STAR-RIS-enabled Integrated Sensing and Communication

Simultaneously transmitting and reflecting reconfigurable intelligent surfaces (STAR-RISs) creates new opportunities for enhancing integrated sensing and communication (ISAC) performance. Large-scale antenna arrays can be deployed to compensate for the severe multiplicative fading associated with STAR-RIS. However, such architectures introduce two practical limitations: acquiring accurate channel state information (CSI) and implementing fully digital beamforming become impractical. To address these challenges, this paper develops a robust and secure transmission scheme for active STAR-RIS-enabled ISAC under imperfect CSI and hybrid beamforming architectures. We jointly optimize the analog beamformer, digital beamformer, and active STAR-RIS coefficients to maximize the minimum secrecy rate. To tackle the resulting non-convex problem, a penalty-based fully digital approximation algorithm is proposed, where the solution is iteratively refined through coordinated update steps. Specifically, the auxiliary fully digital beamformer and active STAR-RIS coefficients are optimized via a majorization– minimization (MM) framework, while the analog and baseband digital beamformers admit closed-form expressions. Numerical results demonstrate that the proposed scheme significantly outperforms passive STAR-RIS benchmarks and achieves performance close to that of fully digital architectures.