Hybrid Beamforming Design for MIMO-ISAC Systems via Consensus-ADMM.

Hybrid analog-digital beamforming has been identified as an attractive technology, since it is able to effectively reduce hardware cost without significantly reducing the system performance, especially in large-scale antenna array systems. In this paper, we consider the problem of hybrid beamforming design for multi-user multiple-input multiple-output integrated sensing and communication (MIMO-ISAC) systems. The problem for the design is formulated as maximizing the minimum transmit energy radiated towards the region-of-interest (ROI) for sensing, subject to the signal-to-interference-plus-noise ratio (SINR) constraints of communication users, unimodular constraints of the analog beamformer phase shifters as well as the total transmission power constraint. To deal with the challenging nonconvex problem, an algorithm based on the consensus alternation direction method of multipliers (CADMM) is developed. Different from the existing two-stage algorithm for such designs, where in general the SINR constraint cannot be satisfied, in the proposed method this constraint is considered during the iterations, so that the SINR requirements can be always guaranteed. Simulation results demonstrate the effectiveness of the proposed algorithm.