用分数模型实现通用纠错码解码，性能优于现有方法。
Score Based Error Correcting Code Decoder
- 将解码转化为连续时间去噪过程，利用神经网络迭代修正观测值。
- 42组测试中39次达到最佳误码率，平均提升0.17dB，最高达0.46dB。
- 支持无需信噪比估计的推理，可灵活调节解码延迟与精度。
纠错码能实现可靠通信，但跨码类和码长的软解码仍具挑战。本文提出SB-ECC，一种基于分数的解码器，将解码建模为连续时间去噪过程。神经去噪器定义了一个概率流常微分方程（ODE），通过奇偶校验约束迭代更新含噪信道观测值，逐步逼近有效码字。模型在无时间/信噪比条件下的多噪声水平上训练，实现推理时无需信噪比估计，并可通过调整ODE求解器预算直接控制解码延迟与精度的权衡。输入使用原始带符号信道观测值以学习连续去噪场。在42个码型/信噪比设置下，SB-ECC在39组中达到最优误码率（BER），平均提升0.17dB，最大提升0.46dB。将求解器从Euler替换为DPM，在保持−ln(BER)基本不变的同时，平均降低8.86%端到端解码时间（最高达12.82%）。
原文摘要 · Abstract (English)
Error-correcting codes enable reliable communication, yet practical soft decoding remains challenging across code families and block lengths. We propose SB-ECC, a score-based decoder that casts decoding as continuous-time denoising. A neural denoiser defines a probability-flow ordinary differential equation (ODE) that iteratively updates the noisy channel observation toward a valid codeword, guided by parity constraints. The model is trained across noise levels without time/SNR conditioning, enabling inference without SNR estimation and supporting a direct latency accuracy trade off controlled by the ODE solver budget. We use the raw signed channel observation as input for learning a continuous denoising field. Across 42 code/SNR settings, SB-ECC achieves the best BER in 39/42 entries, with an average SNR gain of 0.17dB and a maximum gain of 0.46dB over the strongest competing baseline, we showed that swapping the solver from Euler to DPM preserves -ln(BER) while reducing end-to-end decoding time by 8.86% on average (up to 12.82%).
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。