[Paper Review] Low Complexity Iterative Receiver Design for Sparse Code Multiple Access

Fan Wei, Wen Chen|arXiv (Cornell University)|Feb 27, 2020

Advanced Wireless Communication Technologies35 references109 citations

TL;DR

This paper proposes a low-complexity list sphere decoding (LSD) based iterative multiuser detector for SCMA to reduce MPA complexity with negligible performance loss.

ABSTRACT

Sparse code multiple access (SCMA) is one of the most promising methods among all the non-orthogonal multiple access techniques in the future 5G communication. Compared with some other non-orthogonal multiple access techniques such as low density signature (LDS), SCMA can achieve better performance due to the shaping gain of the SCMA codewords. However, despite of the sparsity of the codewords, the decoding complexity of the current message passing algorithm (MPA) utilized by SCMA is still prohibitively high. In this paper, by exploring the lattice structure of SCMA codewords, we propose a low complexity decoding algorithm based on list sphere decoding (LSD). The LSD avoids the exhaustive search for all possible hypotheses and only considers signal within a hypersphere. As LSD can be viewed a depth-first tree search algorithm, we further propose several methods to prune the redundancy visited nodes in order to reduce the size of the search tree. Simulation results show that the proposed algorithm can reduce the decoding complexity substantially while the performance loss compared with the existing algorithm is negligible.

Motivation & Objective

• Motivate SCMA as a promising non-orthogonal multiple access technique for 5G and beyond.
• Exploit the lattice structure of SCMA codewords to reduce decoding complexity.
• Develop a LSD-based MPA detector with node pruning to lower search space.
• Integrate LSD within iterative detection and decoding to improve BER performance.
• Evaluate the trade-off between complexity reduction and performance loss through simulations.

Proposed method

• Model the uplink SCMA system and its factor graph representation.
• Formulate SCMA codewords as sparse multi-dimensional lattice points.
• Apply list sphere decoding (LSD) to perform a depth-first tree search within a hypersphere for each subcarrier.
• Introduce node pruning and a QM/Schnorr-Euchner (SE) enumeration strategy to reduce search complexity.
• Decompose M-PAM signals into a binary sequence to enable consistent LSD application.
• Incorporate LSD within MPA as a list-based approximation to compute extrinsic LLRs.
• Provide an algorithm (Algorithm 1) for LSD-based MPA detection and discuss complexity/parameter T_max trade-offs.

Experimental results

Research questions

• RQ1 Can a list sphere decoding based detector substantially reduce SCMA decoding complexity compared to conventional MPA while preserving performance?
• RQ2 How can lattice structure and node pruning be leveraged to prune the search tree in LSD for SCMA?
• RQ3 What is the impact of integrating LSD with iterative detection and decoding on BER performance in SCMA?
• RQ4 How should constant-modulus and PAM representations be handled to enable efficient LSD in SCMA codewords?

Key findings

• The LSD-based detector substantially reduces decoding complexity with negligible performance loss compared to the standard MPA.
• Sphere decoding is adapted to SCMA by reformulating the problem as a lattice search over collision users on each subcarrier.
• Node pruning, SE enumeration, and a radius-based candidate list (T_max) control complexity while maintaining accuracy.
• A decomposition of M-PAM into BPSK components enables consistent LSD operation for SCMA codewords.
• The algorithm integrates LSD with MPA in an iterative detection and decoding framework to compute extrinsic LLRs.
• Simulation results show marked complexity reduction relative to MPA with minimal BER degradation.