Performance characterization of an SCMA decoder

Characterization of a Sparse Code Multiple Access (SCMA) decoder is performed to measure its error performance and its time and hardware implementation complexity. SCMA is a non-orthogonal technique proposed to support massive connectivity in future 5G wireless telecommunication systems. The reported SCMA decoder is based on the Message Passing Algorithm (MPA). The complexity of SCMA decoding is characterized to identify its implementation bottlenecks. This paper shows how latency and complexity can be reduced with implementations leveraging parallel techniques and high level synthesis. The reported design space exploration allows reducing power and energy consumption.