Neural Network-Based LDPC Decoding for 6G Ultra-Reliable Communications
Dr. Eliya Nachmani, Prof. Yair Be'ery
Tel Aviv University
Abstract
We propose a neural network-enhanced LDPC decoder that achieves near-ML decoding performance for 6G ultra-reliable low-latency communications (URLLC). Our approach uses graph neural networks on the Tanner graph structure of LDPC codes, learning optimal message passing weights that outperform standard belief propagation. The decoder achieves a 0.5 dB gain at 10^-7 block error rate with only 5 iterations (versus 50 for standard BP), enabling the ultra-low latency required for 6G URLLC.
AI Summary
- GNN-enhanced LDPC decoder achieving near-ML performance for 6G URLLC.
- 0.5 dB gain at 10^-7 BLER with only 5 iterations vs 50 for standard BP.
- Graph neural networks learn optimal message passing on Tanner graph.
- Enables ultra-low latency decoding required for 6G URLLC.
Key Findings
- 1Learned weights break symmetries that limit standard BP performance.
- 2The decoder generalizes across different code rates and block lengths.
- 310x fewer iterations directly translate to 10x lower decoding latency.
Industry Implications
Enables 6G URLLC with block error rates below 10^-7.
Demonstrates that AI can improve fundamental communication building blocks.
Could be standardized as an enhanced decoding option in 6G specifications.
Read the Original Paper
Access the full paper on arXiv for complete methodology, results, and references.
Open on arXivRelated Papers
AI-Native Air Interface Design: End-to-End Learning for 6G Physical Layer
University of Stuttgart — 41 citations
AI + Network PapersDigital Twin Networks: AI-Driven Real-Time Network Simulation for 6G
Oulu University / Ruhr University Bochum — 29 citations
AI + Network PapersIntent-Based Network Management with Large Language Models
Universidad Carlos III de Madrid — 16 citations