Multiuser Detection with Compressive Sensing Iterative Reweighed Approach for Grant-Free MIMO-NOMA Systems

In this paper, we focus on the efficient multiuser detection (MUD) problem for multiple-input multiple-output (MIMO) enabled grant-free non-orthogonal multiple access (GF-NOMA) system for massive machine-type communications (MTC). The inherent sparsity of mMTC motivates us to make use of compressive sensing (CS) technology to address the MUD problem. This paper discusses the use of an iterative reweighed (IR) scheme combined with the majorization-minimization (MM) algorithm to recover sparse signals under the Grantfree MIMO-NOMA model. Numerical and simulation results demonstrate that Grant-Free MIMO-NOMA with the proposed IR scheme is capable of reconstructing sparse signals with unknown user activity factors and the proposed algorithm outperforms conventional multiuser detectors