Multiple antenna MPA detection for uplink SCMA systems based on selective Gaussian approximation

In this paper, we propose a selective Gaussian approximation to reduce the complexity of message passing algorithm (MPA) detection for uplink sparse code multiple access (SCMA) systems with multiple antennas. A subset of resource nodes, associated with antennas and subcarriers, are selected and the conventional MPA detection based on maximum likelihood (ML) is applied to those nodes. For the remaining resource nodes, Gaussian approximation is performed instead of ML. We first propose a simple but efficient antenna-subcarrier selection algorithm, and then develop a method to combine the two MPA detection schemes based on ML and Gaussian approximation. Numerical results are presented to show that the proposed MPA detection scheme provides an efficient tradeoff between the performance and receiver complexity.