Loose Beamforming by Antenna Selection in a Multiuser Massive MIMO System

In the 5th generation mobile communication system, implementation of a multiuser massive multiple-input multiple-output system arranging more than 100 elements at a base station has been studied to realize high peak rate and large capacity. On the other hand, with increasing the number of antenna elements used for transmission, the computational load for transmission precoding becomes a serious problem. Loose beamforming which uses one/zero weights, i.e., only selects some useful antenna elements, requires very low complexity but provides poor performance when the element selection fails. In this paper, we evaluate the sum channel capacity performance in three cases of the antenna selection: proposed, random, and genetic algorithm methods in an indoor environment. As a result, the proposed method succeeded in making the loose beamforming by enhancing desired signal power although suppression of interference to other users was still insufficient.