Linear precoding in multi-user massive MIMO systems with imperfect channel state information

Abstract

This paper studies the performance of massive multi-user multiple-input multiple-output (MaMU MIMO) downlink systems. A MaMU system composed by a base station with M antennas and N single-antenna users (M >> N) in a single cell is considered. Several linear precoding techniques, assuming perfect channel state information at the transmitter (CSIT) side, are considered and implemented in order to evaluate their performance. We put special emphasis considering imperfect channel information and its effects over the system performance. The robustness of linear precoding techniques against channel estimation errors and hardware impairments is also evaluated. The impact of the impairments on the system bit error rate (BER) and achievable rate is investigated. Preliminary results show that linear precoders can be employed in massive MIMO systems with promising results. Particularly, the matched filter (MF) precoder, which has the lowest implementation complexity, allows to obtain good results when a large number of antennas are placed at the base station while maintaining constant power.