Learning Enhanced Beamforming Vector From CQIs in 5G NR FDD Massive MIMO Systems: A Tuning-free Approach

In the fifth generation new radio (5G NR) frequency division duplex (FDD) massive multiple-input and multiple-output (MIMO) systems, downlink beamforming relies on the codebook design and limited feedback schemes. Particularly, based on the Type I codebook introduced by 3GPP and the two feedback values (i.e., precoder matrix indicator (PMI) and channel quality indicator (CQI)), base station (BS) aims at acquiring the optimal beamforming vector. Different from previous schemes that focus on the exploitation of PMI, this paper firstly reveals the link between the optimal beamforming vector and CQI, and then devises an intelligent tuning-free algorithm that can learn both the beamforming vector and the associated regularization parameter from the collected CQIs. Numerical results based on channel samples from QUAsi Deterministic RadIo channel GenerAtor (QuaDRiGa) have shown the excellent performance of the proposed algorithm in terms of both beamforming vector acquisition and regularization parameter learning.