Artificial-Noise-Aided Secure Transmit Beamforming for MU-MISO Integrated Sensing and Communication Systems

In this paper, we investigate the physical layer security aspects of integrated sensing and communication (ISAC) systems, where an ISAC base station (BS) communicates with multiple downlink legitimate users and senses a radar target simultaneously. The target of interest is regarded as a potential eavesdropper and there are possibly other eavesdroppers, which attempts to acquire confidential information between BS and legitimate users. We analyze this more practical scenario and employ artificial noise (AN) aided beamforming to ensure secure transmission and sensing performance. Unlike existing research on ISAC multi-user physical layer security, which often employs simplified secrecy metrics such as constrained eavesdroppers signal-to-interference ratio, we consider a more information-theoretic secrecy metric, i.e., achievable sum secrecy rate in this paper. Specifically, we formulate an optimization problem to maximize the sum secrecy rate of the ISAC system by jointly designing beamforming and AN structure, under the constraint of matching error between the designed and desired radar beampatterns. To deal with the challenging non-convex sum rate function, an efficient algorithm based on block coordinate descent and semi-definite relaxation methods are adopted to solve the problem. Numerical simulation results verify the effectiveness of the transmit design and demonstrate a significant secrecy rate while ensuring the required sensing performance.