Deep Learning-Based Blind Multiple User Detection for Grant-free SCMA and MUSA Systems

Abstract

Grant-free random access and uplink non-orthogonal multiple access (NOMA) are techniques to increase the overload factor and reduce transmission latency with signaling overhead in massive machinetype communications (mMTC). Sparse code multiple access (SCMA) and Multi-user shared access (MUSA) are introduced as advanced code domain NOMA schemes. In grant-free NOMA, machine-type devices (MTD) transmit information to the base station (BS) without a grant, creating a challenging task for the BS to identify the active MTD among all potential active devices. In this paper, a novel deep neural network (DNN)-based multi-user detection (MUD) scheme for the grant-free SCMA and MUSA system in an mMTC uplink framework is proposed to jointly identify the received signal's sparsity and the active MTDs in the absence of channel state information. The proposed scheme learns the correlation between the received signal and the multi-dimensional codebook for SCMA and spreading sequences for MUSA schemes and is able to identify the active MTDs from the received signal without any prior knowledge of the device sparsity level. The application of the proposed MUD scheme is further investigated in an indoor factory setting using four different mmWave channels. Numerical results show that when the number of active MTDs in the system is large, DNN-MUD has a significantly higher probability of detection compared to existing approaches over the signal-to-noise ratio range of interest.