Deep Learning for Hybrid Beamforming with Finite Feedback in GSM Aided mmWave MIMO Systems
This work addresses the practical challenge of limited feedback in mmWave MIMO systems, which is crucial for improving spectrum efficiency in wireless communications, though it is incremental by applying deep learning to an existing framework.
The paper tackles the problem of designing hybrid beamformers for GSM-aided mmWave MIMO systems under finite feedback, using deep learning to jointly optimize pilot training, channel estimation, CSI feedback, and beamforming in an end-to-end unsupervised manner. The proposed GsmEFBNet achieves a higher achievable rate with fewer feedback bits compared to conventional algorithms.
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.