Code-aided non-orthogonal multiple access in downlink multiuser MIMO system

Conventional downlink non-orthogonal multiple access (NOMA) uses the power domain for multiple access and applies successive interference cancellation (SIC) to mitigate intra-cluster interference. It requires sophisticated user terminals (UTs) clustering, moreover, imperfect SIC results in error propagation that severely degrades system capacity. In this paper, a novel code-aided multiuser multiple-input multiple-output NOMA (MIMO-NOMA) framework for downlink transmission is proposed. Since the total number of UTs in a cell is in general more than the number of transmit antennas at the base station (BS), purely zero-forcing (ZF) downlink beamforming is unable to mitigate multiuser interference (MUI). We first propose a simple grouping algorithm that separate all UTs in a cell into several clusters, with cluster number less than or equal to BS antennas. ZF-based multiuser beamforming is then employed to remove the inter-cluster interference. To mitigate the intra-cluster interference, we propose to modulate the downlink data for each UT within the same cluster by orthogonal codes, thereby, instead of SIC, matched-filtering can be employed at each UT for interference suppression. Computer simulation results quantify the capacity gain of the proposed coded-aided MIMO-NOMA method over conventional MIMO-NOMA schemes. (C) 2019 The Franklin Institute. Published by Elsevier Ltd. All rights reserved.