Precoding to Reduce Complexity of Factor Graph for MIMO-SCMA

Combining sparse code multiple access (SCMA) with multiple-input multiple-output (MIMO) can improve spectral efficiency of SCMA. However, in a downlink MIMO-SCMA system, each user receives signals from all transmit antennas. This increases the computational complexity of detection using the message passing algorithm (MPA) to estimate the desired symbols. To solve this problem, we propose a precoding scheme that can reduce the computational complexity of MPA without compromising the bit error rate (BER) performance. In the proposed scheme, weights are multiplied by symbols to be transmitted by the base station so that each user receives signals with a reduced number of superposed symbols. Computer simulation results show that the proposed scheme can improve the BER performance while reducing the computational complexity of MPA.