Adaptive Downlink Beamforming based on SRS for Channel Estimation using Coherence Bandwidth Characteristic in sub-6GHz 5G NR

Abstract

The precise channel estimation is a prerequisite to acquire beamforming performance that the massive multi-input and multi-output (MIMO) system guarantees. There is pre-coding matrix indicator (PMI) based downlink beamforming as the beamforming method using downlink channel estimation. The PMI based downlink beamforming uses grid of beam (GoB) based on codebook. That is why it is challenging to track down the exact location of a mobile station and to obtain the maximum beamforming gain. The sounding reference signal (SRS) based downlink beamforming using uplink channel estimation gets the correlation coefficient between SRSs received from each receiver of a base station and obtains the maximum beamforming gain by tracking down the location of a mobile station accurately. However, as the SRS based downlink beamforming is required to be detected at each receiver of a base station, it has coverage limitation at the cell edge compared to the PMI based downlink beamforming using the combined feedback at a base station. This paper proposes that system obtains the maximum beamforming gain by using the SRS based downlink beamforming at the cell center and supplements coverage limitation by switching it to the PMI based downlink beamforming at the cell edge and using coherence bandwidth.