Hybrid Beamforming Design with Finite Resolution Phase Shifters in Multi-User Systems

Hybrid beamforming is a key technology in the next generation wireless communication. It achieves similar performance as full digital beamforming with much reduced power consumption and hardware complexity. Compared with fully-connected hybrid beamforming architecture, the partially-connected architecture is of particular interest as it further simplifies the hardware design. In this work we study the partially-connected beamforming with finite resolution phase shifters in downlink massive MIMO systems. Fairness or the quality of service (QoS) of each user is considered. In this prob-lem, partially-connected architecture and finite resolution phase shifters put additional constraints to the system. We propose an efficient penalty-based algorithm to jointly optimize the digital and analog beamforming, which yields analog beamforming in discrete values. Numerical results show that the proposed algorithm performs better than other benchmark algorithms. It achieves satisfactory performance with low resolution phase shifters.