Deep Learning Approach for Improving Spectral Efficiency in mmWave Hybrid Beamforming Systems

초록

Hybrid beamformer design plays an important role in millimeter wave multiple input multiple output systems. In this paper, we propose a deep learning (DL) neural network for hybrid precoders and combiners to improve spectral efficiency. With the received signal and channel matrix as the input, the proposed DL network estimates the beamformer matrix as output. The proposed DL approach does not require prior knowledge such as angle features and channel information. Thus, it provides improved spectral efficiency compared to non-DL approaches.

키워드