User Scheduling Schemes of DFT-based Hybrid Beamforming Multiuser Systems

This paper investigates the achievable rate of a multiuser multiple-input multiple-output system with a hybrid architecture based on the discrete Fourier transform (DFT) processing under Rayleigh fading condition. We first derive the lower bound of the achievable rate when the baseband processing is performed by the minimum mean-square error (MMSE) receiver and the base station (BS) has perfect channel state information (CSI). Analytical results show that the achievable rate is proportional to the large-scale fading coefficients and can be improved by increasing the number of radio frequency (RF) chains. Based on the results, a novel proportional fair (PF) user scheduling scheme is proposed, which updates the priority of users during the scheduling. Simulations show that the theoretical bound is very tight with the Monte-Carlo and validate that the proposed scheduling scheme not only achieves higher rate than round robin (RR) selection, but also has a significant advantage in scheduling fairness.