A Reduced Overhead Low Complexity Sum Rate Optimization Algorithm for Massive MIMO System

In this paper, we study a joint spatial division multiplexing (JSDM) beamforming scheme based on the weighted minimum mean square error (WMMSE), aiming to optimize the sum rate of FDD massive MIMO system with low complexity and reduced-dimensional channel state information at the transmitter (CSIT). The proposed algorithm concatenates a prebeamforming matrix, which depends on the channel second-order statistics and is achieved by block diagonalization (BD), with a multi-user precoder to maximize the sum rate of the system by the low complexity WMMSE algorithm. The simulation results show that the WMMSE-based algorithm effectively increases the sum rate of the system comparing to the conventional JSDM which designs the two-stage beamformer with BD and zero forcing method. At the same time, the effect of effective rank in BD on the performance of the algorithm is analyzed.