Efficient Message Passing Receivers for Downlink MIMO-SCMA Systems

Hao Cheng,Chuan Zhang,Yongming Huang,Luxi Yang

DOI: https://doi.org/10.1109/tvt.2022.3151936

IF: 6.8

2022-01-01

IEEE Transactions on Vehicular Technology

Abstract:Sparse code multiple access (SCMA) and multiple input multiple output (MIMO) are considered as two efficient techniques to provide both massive connectivity and high spectrum efficiency for future machine-type wireless networks. However, the computational complexity of the conventional message passing algorithm (MPA) for the downlink MIMO-SCMA detection increases exponentially with the degree of variable nodes (VNs). To address this issue, two novel fixed low complexity MPA detectors are proposed by utilizing the sparse feature of codewords, which are based on the serial schedule strategy in order to accelerate their convergence. First, a maximum distance MPA (MDMPA) detector is introduced to reduce the number of VNs involved in the message updating procedure in the first $T_0$ iterations and fix $TJ_0$ codewords at the $L_0$ iteration, which subsequently drops the redundant codeword combinations in the remaining iterations. To further improve the bit error ratio (BER) performance, an improved MDMPA (IMDMPA) is proposed, in which, the VNs associated with the target user are always allowed to participate in the iterative propagation of message. An efficient linear squares (LS) channel estimation scheme is also derived based on block pilots by stacking all the received signals at the same subcarrier together. Simulation results illustrate that the proposed MDMPA and IMDMPA detectors significantly reduce the computational complexity of the original MPA, but with comparable decoding capabilities, and the proposed LS channel estimator can closely approach the corresponding lower bound over frequency selective channels.