Energy efficient phase interpolator based hybrid beamforming architecture for massive MIMO system

Massive MIMO embraces digital beamforming techniques to direct power to the intended users. However, digital beamforming is power-consuming as each antenna has a dedicated RF chain. To address this issue, a hybrid beamforming architecture was proposed, where analog beamforming is combined with a low dimension digital beamforming. Consequently, the number of RF chains was significantly reduced. The conventional hybrid beamforming architecture deploys phase shifters in analog beamforming, which consumes large power as the number of antennas increases. This work proposes a phase interpolator-based analog beamforming to reduce the number of phase shifters in the fully connected hybrid beamforming architecture. Additionally, we have formulated an optimization problem to search for the best analog and digital precoding in order to maximize spectral efficiency. Solving this problem, we have proposed a tabu search (TS) based algorithm, which uses a genetic algorithm crossover feature to search for the new neighbor. Simulation results show that the proposed method achieves high spectral efficiency and energy efficiency. Specificall, when SNR = 0 dB, the proposed method achieves 95% spectral efficiency of an optimal solution, and achieves 20% higher energy efficiency than conventional method. Furthermore, the proposed method achieves high spectral efficiency with low computational complexity, in comparison to conventional methods.