Impact of Self-Equalization in a Spectral Efficiency Analysis in Massive MIMO

Abstract

This work evaluates the impact of self-equalization in a Massive Multiple-Input Multiple-Output (MIMO) system uplink in which the performance of the Filter Bank Multicarrier (FBMC) and the Orthogonal Frequency Division Multiplexing (OFDM) are compared in terms of spectral efficiency (SE) under a single-cell environment. The FBMC is referred as one of the strongest candidates to replace OFDM in 5G networks, but it is greatly affected by noise, inter-user interference, and distortion caused by the channel frequency selectivity in a MIMO channel. However, a phenomenon referred to as “self-equalization” occurs when the channel employs a larger ratio between the number M of base station (BS) antennas and the number K of users in Massive MIMO (MM), turning the effects of the channel frequency selectivity negligible. Under that condition, the FBMC presents better SE than OFDM mostly due to the absence of the cyclic prefix. Thus, by adopting a threshold in terms of the output mean squared error of the estimated transmitted symbols in MM, the results indicate the viability of FBMC, showing that although it requires a larger M/K ratio than OFDM to start working, after surpassing the threshold it presents a much higher SE which clearly justifies the extra number of BS antennas.