Soft-Output GRAND-Aided Macrosymbol

2026-09-27

AI Summary

This study addresses the performance bottleneck in non-orthogonal multiple access (NOMA) systems caused by the incompatibility between hard-decision detection and soft decoders. To overcome this limitation, we propose the SOGRAND-AM framework, which builds upon the guessing random additive noise decoding (GRAND) technique to perform joint multi-user detection and decoding at the macro-symbol level. The core innovation lies in generating calibrated bit-level posterior probabilities to support soft-decision error correction, thereby transcending conventional hard-output constraints and enabling transparent integration with existing soft-input decoders. Experimental results demonstrate that, compared to hard-input baselines, the proposed framework improves the bit error rate performance of the outer soft decoder by approximately 2 dB, significantly enhancing the overall error-correction capability of the system.

Abstract

Guessing Random Additive Noise Decoding-Aided Macrosymbol (GRAND-AM) provides a coding-based solution for non-orthogonal multiple access (NOMA) systems, enabling joint multiuser detection and error correction. GRAND-AM outperforms prior methods in symbol error rates while providing hard-detection output. We introduce soft-output GRAND-AM (SOGRAND-AM), a macrosymbol-level joint multiuser detection and decoding framework which generates calibrated bitwise a posteriori error probabilities. SOGRAND-AM extends GRAND-AM by enabling soft-decision outer forward error correction decoders, transparently integrating with existing decoders. SOGRAND-AM achieves approximately 2 dB improvement in bit error rate at the output of the outer soft-input decoder compared to hard input with GRAND-AM.

Problem

• Non-Orthogonal Multiple Access (NOMA)
• Soft-Output Decoding
• Joint Multiuser Detection
• Macrosymbol Decoding
• GRAND

Innovation

• Soft-output GRAND-AM
• Non-orthogonal multiple access (NOMA)
• Joint multiuser detection and decoding
• A posteriori error probabilities
• Soft-decision forward error correction