An Efficient Hybrid Diagonalization for Multiuser mmWave Massive MIMO Systems

In millimeter wave (mmWave) massive multiple-input multiple-output (MIMO) systems, hybrid beamforming structure can be used to improve spectral efficiency by utilizing less radio-frequency (RF) chains. In this paper, we consider the downlink communication of a mmWave massive MIMO system and propose a novel hybrid beamforming algorithm based on extended block diagonalization (BD) and equal gain transmission (EGT) method with near optimal performance and low complexity. At the RF analog domain, we harvest the large array gain by using the EGT method and discrete Fourier transform (DFT) matrix. At the baseband digital domain, the extended BD algorithm is performed. The extended BD algorithm considers both the interference null space as well as the characteristic of the user's signal space while the traditional BD algorithm only considers the interference null space. Then we also analyze the performance of the proposed algorithm in different channel models. According to the simulation results, we find that the proposed algorithm performs well both in sparse channels and Rayleigh channels. To the best of author's knowledge, the performance of the proposed algorithm is better than other low complexity hybrid beamforming algorithms.