Radar Mutual Information Maximization for Movable Antenna-Enabled ISAC Systems

This paper jointly optimizes transmit beamforming and antenna positions for an integrated sensing and communication (ISAC) system equipped with movable antennas (MAs). We consider a downlink dual-functional radar-communication system, where a base station (BS) serves a group of communication users while detecting multiple targets. The objective is to maximize the radar mutual information (MI), subject to the transmit power, communication signal-to-interference-plus-noise-ratio (SINR) and MA geometry constraints. This problem is challenging due to the strong non-convexity of the objective and tightly coupled variables.We derive an asymptotically tight upper bound for the radar MI, and develop an efficient alternating optimization algorithm that applies semidefinite relaxation (SDR) for beamforming and successive convex approximation (SCA) for antenna positions. The optimality of the beamforming is ensured by retaining the rank-one property after SDR. Simulations show that our algorithm offers a substantial gain in sensing capability, improving radar MI considerably over benchmarks.