Fast Specific Absorption Rate Aware Beamforming for Downlink SWIPT via Deep Learning

This paper investigates fast implementation of the optimal transmit beamforming design for simultaneous wireless information and power transfer in the multiuser multiple-input-single-output downlink with specific absorption rate (SAR) constraints. The problem of interest is to maximize the received signal-to-interference-plus-noise ratio and the energy harvested for all receivers, while satisfying the transmit power and the SAR constraints. The optimal solutions can be obtained via convex optimization and bisection search but with high complexity. To reduce the computational complexity, this paper proposes the deep learning technique to predict key features of the problem and then recover the beamforming solutions with much reduced complexity. Simulation results demonstrate that our proposed algorithms can significantly reduce the algorithm execution time while maintaining satisfactory performance.