A multi-user linear equalizer for uplink broadband millimeter wave massive MIMO

Abstract

Most previous work on hybrid beamforming millimeter wave (mmWave) communications and massive multiple input multiple output (MIMO) systems focused on narrowband channels. However, it is expected that the mmWave channels will be wideband, and for that reason, it is important to design solutions for frequency-selective channels. Therefore, in this paper, we consider the design of a hybrid analog/digital multiuser linear equalizer for broadband mmWave massive MIMO systems, optimized using the bit error rate (BER) as a metric. For the digital part, a closed-form solution is obtained by showing that the minimum mean square error (MMSE) and minimum BER solutions are identical. The analog part, which is assumed to be the same for all subcarriers, is computed iteratively over the radio frequency (RF) chains, and it is shown that the solution obtained in a previous iteration may be updated inexpensively to obtain a solution for the current iteration. The simulation results show that the performance gap between the proposed hybrid broadband multiuser equalizer and full digital equalizer decreases monotonically and exponentially with the number of RF chains until reaching zero. Moreover, the gap decay rate is faster for sparser mmWave channels, where fewer RF chains are required to achieve the same performance of the full digital architecture.