An Energy-Efficient Hybrid Precoding Algorithm for Multiuser mmWave Massive MIMO Systems

Millimeter wave (mmWave) systems with large number of antennas are considered as an enabling technology for the fifth generation (5G) cellular networks. For such systems, the traditional fully digital precoding is impractical due to the high cost of radio frequency (RF) chains. To overcome this difficulty, hybrid precoding (HP) is considered. However, in the conventional HP architecture, the number of analog phase shifters (APSs) and RF adders increases linearly with respect to the number of transmit antennas, leading to considerable energy consumption. In this paper, an energy-efficient weighted minimum mean square error (WMMSE) based hybrid precoding algorithm is proposed to maximize the achievable system sumrate. Through a novel adaptive connection network, a partially-connected HP structure is employed to implement the design of the RF precoder, where each RF chain is connected to only a part of antennas. Given the RF precoder and combiner, we can optimize the baseband precoder and combiner by using WMMSE algorithm. Numerical results show that the proposed algorithm can achieve near optimal sum-rate performance and significantly improve the energy efficiency as compared to the conventional HP algorithm.