Hybrid Beamforming for Utility Maximization in Multiuser Broadband Millimeter Wave Systems

Hybrid beamforming (HBF) has been considered as a key technology in millimeter wave (mmWave) massive multiple-input multiple-output (MIMO) communication systems, since it can significantly reduce the number of radio frequency chains, and in turn the hardware cost and power consumption, while providing performance comparable to that of a fully-digital beamforming architecture. Conventional HBF designs mostly focused on the spectral efficiency maximization for a certain HBF connection architecture such as fully-connected or partially-connected one. In this article, we investigate HBF optimization with the objective of sum-utility maximization (SUM) for multi-user broadband mmWave MIMO systems with arbitrarily connected architectures. In particular, we show that the SUM HBF optimization can be equivalently converted to a specific weighted mean square error minimization (WMMSE) problem, which leads to a convenient algorithmic approach to directly tackle the difficulty of the SUM problem. Furthermore, by utilizing the alternating minimization and manifold optimization methods, we propose an efficient iterative algorithm to solve the non-convex WMMSE problem and prove its convergence. Simulation results demonstrate the fast convergence as well as the significant performance improvement of the proposed algorithm over the conventional ones.