Secure Hybrid Beamforming Design for mmWave Integrated Sensing and Communication Systems

This paper investigates secure hybrid beamforming with subarray architectures in a millimeter-wave (mmWave) integrated sensing and communication (ISAC) system. To achieve secure communication against a potential eavesdropper, we design the sensing signal in a dual-functional form, serving both to detect the target and to generate artificial noise (AN) that disrupts the eavesdropper. With fixed subarray architectures, the security and sensing performance are analyzed, including the impact of imperfect channel state information (CSI). Moreover, the design of analog and digital beamformers is formulated as the problem of minimizing transmit power while satisfying the constraints of sensing beampattern and secrecy rate. To solve the non-convex problem, semi-definite relaxation (SDR) and successive convex approximation (SCA) techniques are employed to transform the original problem and obtain the solution. For the scenario with imperfect CSI, the robust hybrid beamforming is designed based on S-procedure. Additionally, to optimally utilize spatial domain resources provided by multiple antennas, a dynamic subarray architecture is introduced, along with two dynamic subarray partitioning algorithms. This flexible architecture can be dynamically adjusted according to the CSI, ensuring a bidirectional alignment with the actual channels and the proposed scheme. Specifically, the greedy algorithm and the secrecy rate increment maximization (SRIM) algorithm are designed to partition the dynamic subarray, and the alternating optimization is utilized to further improve the quality of the solution. Numerical results demonstrate the effectiveness of the proposed schemes.