A Novel SRS-DMRS Unified Channel Estimation-Based Enhanced Downlink Beamforming in TDD Multiuser Massive MIMO-OFDM System

Abstract

This paper proposes a novel dynamic beamforming (BF) technique exploiting unified channel estimation obtained from uplink demodulation reference signal (UL-DMRS) and sounding reference signal (SRS). More specifically, the conventional SRS-based dynamic BF, typically used for reciprocitybased downlink (DL) BF in time-division duplexing (TDD)-based multiple-input multiple-output (MIMO) systems, is enhanced by an opportunistic unification of SRS-based and UL-DMRS-based channel estimates. The performance of SRS-based BF is highly dependent on the accuracy of channel state information (CSI) available at the base station, which poses a significant challenge in the presence of channel aging. To address this, our proposed unified BF technique repurposes the UL-DMRS to enhance the CSI accuracy of SRS-based DL BF without increasing additional pilot overhead. Since UL-DMRS transmission is tied to UL data transmission schedule and does not have the flexibility like SRS, we introduce an opportunistic construction of the unified channel estimation matrix based on UL activity. Extensive system-level simulation results based on 5G new radio (NR) sub-6 GHz massive MIMO framework demonstrate that the proposed unified beamforming significantly outperforms the conventional SRSbased BF by better mitigation of channel aging.