Low Complexity Hybrid Precoder Design for mmWave Multi-User MIMO Systems: A Non-Iterative Approach

Abstract

Massive multiple-input multiple-output (MIMO) will be an important ingredient in millimeter-wave (mmWave) cellular communication systems. A natural application of massive MIMO is simultaneous transmission to multiple users. Unfortunately, the hardware constraints in massive MIMO make it difficult to apply the conventional fully digital beamforming techniques, especially in the mmWave bands. Thus hybrid beamforming (HB) in which the overall beamformer consists of a low-dimensional digital beamformer followed by an analog beamformer has been proposed for reducing the number of costly radio frequency (RF) chains in massive MIMO systems. This paper considers an HB design for multi-user massive MIMO systems. We design a sub-optimal analog precoder based on the downlink-uplink duality. To support the general situation where the number of RF chains is not necessarily equal to the number of users, we propose a greedy selection algorithm to allocate RF chains. Finally, we consider the hardware impairment in analog part and extend the proposed algorithm to support the use of finite resolution phase shifters. Simulation results demonstrate that the proposed HB method makes a better tradeoff between complexity and performance compared with existing methods.