This letter addresses the issues of gradient vanishing and performance degradation in deep-unfolded neural LDPC decoders. We introduce a residual-based and weight-shared neural normalized min-sum (RSNNMS) decoder, which incorporates residual connections into the shared neural NMS (SNNMS) architecture to actively leverage messages from prior iterations, thereby enhancing current layer updates and stabilizing training in deep networks. Observing that the residual weights learned by RSNNMS exhibit varying contributions across different layers, we further propose a threshold-controlled variant (TRSNNMS) that applies a threshold during inference obtained on the validation data set to selectively enable residual connections based on their significance, thereby reducing computational complexity while further improving decoding performance. Simulation results for an LDPC code with block length 576 and code rate 0.75 demonstrate that, at a bit error rate (BER) of $10^{-6}$ , the RSNNMS decoder achieves approximately 0.26 dB signal-to-noise ratio (SNR) gain over the SNNMS decoder, and the TRSNNMS decoder yields an additional 0.22 dB gain over RSNNMS.
In deep-space communications, conventional low-density parity check (LDPC) decoding algorithms suffer from slow convergence and high latency due to inefficient message scheduling. While residual-based layered belief propagation (RB-LBP) improves convergence, its serial scheduling limits hardware parallelism. Similarly,...
This work proposes a Transformer-based decoder for convolutional codes designed for soft-decision decoding over AWGN and fading channels, which serves as an efficient decoding scheme for convolutional codes with various code rates and proposes a Bidirectional Context Aggregation mechanism to enable the model to fully c...
Guang-Ya Hu, Tian-Yi Li, Chao-Ping Xing· Proceedings of the Thirty-Fi...· 0 citations
High-speed optical fiber communication systems require forward error correction decoders that combine strong error correction with low latency and efficient implementation. Generalized low-density parity-check (GLDPC) codes improve finite-length performance and lower the error floor, but their generalized check (GC) no...
Chang-Feng Wang, Rui Jiang, Xin-Ke Tang et al.· Applications of Machine Lear...· 0 citations
The Error Correction Code Transformer (ECCT) is a seminal state-of-the-art model-free neural decoder for linear block codes, leveraging parity-check-matrix-aware masked self-attention to exploit algebraic code structure. A well-known limitation of deep Transformer architectures trained with Pre-Layer Normalization (Pre...
Elaheh Bassak, Elvino S. Sousa· IEEE International Symposium...· 0 citations
Neural channel decoders have attracted growing attention as promising routes for high-performance decoding of short block codes. A recent milestone is denoising diffusion error correction codes (DDECC), which fuses diffusion probabilistic models with a Transformer-based denoiser and outperforms earlier neural approache...
Jian Zheng, Jia-Huan Ni, Zong-Yao Li et al.· IEEE Transactions on Cogniti...· 0 citations
ZeroCode addresses diverse code-design requirements within a unified framework, providing solutions with optimized decoding performance under given constraints, and offers trade-offs between decoding performance and complexity without retraining.
Ju-Hyeong Lee, Yongjune Kim, Sang-Hyo Kim et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
