Spectral Efficiency of DFT-Based Processing Hybrid Architectures in Massive MIMO

Abstract

This letter investigates the achievable spectral efficiency (SE) of massive multiple-input multiple-output transmission systems with hybrid architectures based on discrete Fourier transform (DFT) processing, where the base station (BS) has perfect channel state information and baseband processing is performed by zero-forcing precoding. We derive tractable upper and lower bounds on the achievable SE. Based on these results, the effects of the number of radio frequency (RF) chains, signal-to-noise ratio (SNR), and the number of users are revealed. Compared to hybrid architectures with ideal and quantized phase shifters, simulations indicate that the achievable SE with DFT processing is inferior to the one with ideal phase shifters. In addition, the achievable SE with DFT processing is independent of the number of BS antennas and can be improved by increasing the SNR and the number of RF chains, while there exists an optimal number of users that maximizes the total achievable SE.