Modified Sphere Decoding for Sparse Code Multiple Access

Sparse code multiple access (SCMA), a novel non-orthogonal multiple access scheme for fifth generation wireless, has drawn much attention. Since optimal SCMA detection incurs high computational complexity, we develop a modified single tree search (MSTS) to obtain soft outputs for coded SCMA. The MSTS algorithm incorporates the sorting of the channel matrix and computing a non-zero low bound before searching in order to reduce complexity during the search stage. The simulation results show that MSTS not only has lower computational complexity than detectors based on the message passing algorithm (MPA) and repeat tree search but also keeps its performance level close to that of MPA.