Deep Learning for Hybrid Beamforming with Finite Feedback in GSM Aided mmWave MIMO Systems

TL;DR

This paper proposes a deep learning-based joint optimization framework for hybrid beamforming in GSM aided mmWave MIMO systems, addressing practical issues like finite feedback and imperfect CSI to improve spectral efficiency.

Contribution

It introduces GsmEFBNet, a multi-resolution deep learning network for end-to-end hybrid beamforming with reduced feedback in GSM mmWave MIMO systems, a novel approach compared to existing methods.

Findings

GsmEFBNet achieves higher achievable rates.

Fewer feedback bits are needed for comparable performance.

Deep learning enhances practical beamforming in mmWave MIMO.

Abstract

Hybrid beamforming is widely recognized as an important technique for millimeter wave (mmWave) multiple input multiple output (MIMO) systems. Generalized spatial modulation (GSM) is further introduced to improve the spectrum efficiency. However, most of the existing works on beamforming assume the perfect channel state information (CSI), which is unrealistic in practical systems. In this paper, joint optimization of downlink pilot training, channel estimation, CSI feedback, and hybrid beamforming is considered in GSM aided frequency division duplexing (FDD) mmWave MIMO systems. With the help of deep learning, the GSM hybrid beamformers are designed via unsupervised learning in an end-to-end way. Experiments show that the proposed multi-resolution network named GsmEFBNet can reach a better achievable rate with fewer feedback bits compared with the conventional algorithm.