Variational Diffusion Channel Decoder

2026-05-17

AI Summary

This work proposes an efficient neural channel decoding architecture that integrates belief propagation (BP) with a variational diffusion model, embedding the domain knowledge–driven BP process into a diffusion-based generative framework for the first time. While existing neural decoders achieve strong error-correction performance, they typically suffer from large model sizes and high computational complexity, hindering deployment in resource-constrained real-time systems. The proposed method substantially reduces both model scale and inference overhead while maintaining near-optimal decoding performance. It outperforms current state-of-the-art neural decoders by effectively balancing high error-correction capability with practical deployability.

Abstract

Neural channel decoder, as a data-driven channel decoding strategy, has shown very promising improvement on error-correcting capability over the classical methods. However, the success of those deep learning-based decoder comes at the cost of drastically increased model storage and computational complexity, hindering their practical adoptions in real-world time-sensitive resource-sensitive communication and storage systems. To address this challenge, we propose an efficient variational diffusion model-based channel decoder, which effectively integrates the domain-specific belief propagation process to the modern diffusion model. By reaping the low-cost benefits of belief propagation and strong learning capability of diffusion model, our proposed neural decoder simultaneously achieves very low cost and high error-correcting performance. Experimental results show that, compared with the state-of-the-art neural channel decoders, our model provides a feasible solution for practical deployment via achieving the best decoding performance with significantly reduced computational cost and model size.

Problem

• neural channel decoder
• computational complexity
• model storage
• error-correcting capability
• resource-sensitive systems

Innovation

• variational diffusion model
• neural channel decoder
• belief propagation
• error-correcting performance
• computational efficiency