Learning to Decode Linear Codes with Generalized Belief Propagation Network
arXiv cs.AI · 5 Nov 2026, 04:29 AEST
A new Generalized Belief Propagation Network (GBPNet) has been proposed to enhance the decoding of linear codes. This architecture shows improved performance over traditional methods and state-of-the-art neural decoders.
Key facts
- Error correcting codes are a fundamental component of communication systems, enabling reliable data transmission over noisy channels.
- Recently, neural decoders have demonstrated notable performance gains over classical iterative algorithms.
- The architecture uses cascaded variable-node updates and learned element-wise message scaling while retaining the Tanner graph structure.
- We also introduce two training strategies, progressive training and high-SNR training, that improve decoding accuracy.
- Experimental results on several linear codes show that GBPNet achieves lower bit error rates than conventional belief propagation and the evaluated state-of-the-art neural decoders.
Read aloud
A new Generalized Belief Propagation Network GBPNet has been proposed to enhance the decoding of linear codes. This architecture shows improved performance over traditional methods and state-of-the-art neural decoders. Error correcting codes are a fundamental component of communication systems, enabling reliable data transmission over noisy channels. Recently, neural decoders have demonstrated notable performance gains over classical iterative algorithms. The architecture uses cascaded variable-node updates and learned element-wise message scaling while retaining the Tanner graph structure. We also introduce two training strategies, progressive training and high-SNR training, that improve decoding accuracy. Experimental results on several linear codes show that GBPNet achieves lower bit error rates than conventional belief propagation and the evaluated state-of-the-art neural decoders.
Sources in this cluster
- Learning to Decode Linear Codes with Generalized Belief Propagation Network
arXiv cs.AI · 5 Nov 2026, 04:29 AEST