Machine Learning-Based Beamforming in Two-User MISO Interference Channels

Abstract

As the demand for data rate increases, interference management becomes more important, especially in small cell environment of emerging wireless communication systems. In this paper, we investigate the machine learning-based beamforming design in two-user MISO interference channels. To see the possibilities of machine learning in beamforming design, we consider simple beamforming, where each user chooses one between two popular beamforming schemes, which are the maximum ratio transmission (MRT) beamforming and the zero-forcing (ZF) beamforming. We first propose a machine learning structure that takes transmit power and channel vectors as input and then recommends two users' choices between MRT and ZF as output. The numerical results show that our proposed machine learning-based beamforming design well finds the best beamforming combination and achieves the sum-rate more than 99.9% of the best beamforming combination.