A Novel Efficient Deep Unfolding Algorithm Based on WMMSE for Downlink Beamforming
Jinxing Yang, Yuting Cai · 2023
Downlink beamforming has received great attention in wireless communications. Although the popular weighted minimum mean square error (WMMSE) algorithm can provide excellent performance, it exhibits prohibitively computational complexity. To address this problem, based on the structure of WMMSE algorithm, we propose an efficient ahead momentum-based deep unfolding neural network (AMDUNN) algorithm. Simulation results verify that the proposed AMDUNN remarkably outperforms the existing deep unfolding algorithms with lower complexity.