Score Based Error Correcting Code Decoder
For communication systems requiring reliable decoding, SB-ECC provides a novel score-based approach that outperforms existing decoders across many code families and noise levels.
SB-ECC casts error-correcting code decoding as continuous-time denoising, achieving the best BER in 39/42 code/SNR settings with an average SNR gain of 0.17dB over the strongest baseline, and reducing decoding time by up to 12.82% with a different ODE solver.
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%).