Efficient Beamforming Design for Cell-Free Integrated Sensing and Communication Systems

Cell-free massive MIMO systems enhance connectivity, user throughput, and spectral efficiency in ISAC systems through intelligent beamforming design. Existing optimization methods for beamformers are complex and increase fronthaul loads. To address this, we propose an unsupervised learning algorithm for designing communication and sensing beamformers in cell-free ISAC systems. Our leader-follower training approach balances the maximization of sensing signal-to-noise ratio (SSNR) and signal-to-interference-plus-noise ratio (SINR). This decen-tralized scheme reduces central processing and fronthaul link loads. Two leader models, each focused on one task, guide a follower model to balance both metrics. Our approach achieves performance comparable to state-of-the-art solutions while being at least three orders of magnitude more computationally efficient.