Being an effective non-orthogonal multiple access (NOMA) technique, sparse code multiple access (SCMA) is promising for future wireless communication. Compared with orthogonal techniques, SCMA enjoys higher overloading tolerance and lower complexity because of its sparsity. In this paper, based on deterministic message passing algorithm (DMPA), algorithmic simplifications such as domain changing and probability approximation are applied for SCMA decoding. Early termination, adaptive decoding, and initial noise reduction are also employed for faster convergence and better performance. Numerical results show that the proposed optimizations benefit both decoding complexity and speed. Furthermore, efficient hardware architectures based on folding and retiming are proposed. VLSI implementation is also given in this paper. Comparison with the state-of-the-art have shown the proposed decoder's advantages in both latency and throughput (multi-Gbps).Index Terms-Sparse code multiple access (SCMA), deterministic message passing algorithm (DMPA), folding, retiming, VLSI.• Error propagation: For SIC, if an error occurs, all users afterward are likely to be decoded incorrectly.• Decoding latency: User power sorting is involved in SIC, and causes good overhead latency compared to other methods. Since the data with the lowest power is decoded last, the latency will even higher. Therefore, SCMA employs MUD instead of SIC. Thanks to its sparsity, message passing algorithm (MPA) can be applied for better decoding performance.
B. Sparse Code Multiple AccessSCMA was proposed in 2013 [15], trying to increase user scale via a new perspective: enabling more efficient multiple access by non-orthogonal sparse spreading codes of users.1) Properties of SCMA: As a promising MA, SCMA has the properties: i) multiplexing in frequency domain; ii) codebook based on both mapping and spreading; iii) multidimensional constellation for shaping gain and spectral efficiency; iv) non-orthogonality ensuring more accessed users; v) spreading which reduces noise interference and enhances system robustness; and vi) sparsity which reduces decoding complexity. Thanks to these properties, SCMA is more physically realizable and overloading tolerant, compared to other MAs [16]. Details of SCMA can be found in Section II.2) Challenges of SCMA:• Throughput: Though the throughput of SCMA outperforms other MAs, especially orthogonal ones, it is hard to