SECS: A Soft Ensemble-Combining Stage for Low-Latency Decoding
2026-08-10 • Information Theory
Information Theory
AI summaryⓘ
The authors study a way to speed up decoding in communication systems by running multiple simple decoders at once. They find that these simple decoders are not great at fixing errors alone but are good at ranking which bits are most reliable. Using this insight, they create a method that combines early results from several decoders and then uses a re-encoding step to find a very good estimate of the sent message faster than usual. They tested this approach on a few different codes and showed it gets close to the best possible decoding while requiring much less computation.
ensemble decodingbelief propagationreliability orderingsoft combiningextrinsic messagesordered-statistics decodingmaximum likelihood decodingBose-Ray-Chaudhuri-Hocquenghem codescyclic codesdecoding latency
Authors
Felix Krieg, Paul Bezner, Stephan ten Brink
Abstract
Ensemble decoding is a promising technique for ultra-reliable low-latency communication, as it trades hardware parallelism for decoding latency by running M diverse belief propagation (BP) decoders in parallel. Conventional ensembles select their output from the candidates. Hence, decoding fails whenever no member finds the correct codeword. In this work, we show that shallow, diverse BP members are poor correctors, but excellent sorters. Based on this observation, we propose a soft ensemble-combining stage (SECS) that combines extrinsic messages after only a few iterations, yielding a reliability ordering whose most reliable positions are nearly error-free. A subsequent re-encoding stage (like ordered-statistics decoding (OSD)) converts this ordering into a near-maximum likelihood (ML) candidate codeword, at a fraction of the BP latency of a fully converged ensemble. We demonstrate the proposed scheme on three short codes: a (63,30) Bose-Ray-Chaudhuri-Hocquenghem code, an overcomplete PG(2,8) code, and a search-designed (105,53) cyclic code. In all cases, the SECS with OSD post-processing closes most of the gap to ML decoding, while significantly reducing the number of required BP iterations.