Cost-Aware Neural Early Stopping for Local Constraint OSD Decoders

TL;DR

This paper introduces a neural early stopping protocol for local constraint OSD decoders that reduces computational cost by predicting when to stop searching, maintaining performance across various SNRs.

Contribution

It presents a cost-aware neural early stopping method for LC-OSD decoders, enabling explicit control over search effort and error risk, with a single model effective across SNRs.

Findings

Significant reduction in TEP count with minimal FER degradation.

Effective across multiple code families and SNR ranges.

Uses a single global model for all operating conditions.

Abstract

Local constraint ordered statistics decoding (LC-OSD) provides strong soft decision performance for short block length linear codes, but its practical cost is dominated by the number of tested error patterns (TEPs). This paper proposes a neural early stopping (NES) protocol for LC-OSD with explicit cost control through one trade-off parameter balancing frame error risk and search effort. The proposed approach is trained with frame error rate (FER)-aligned supervision at predefined checkpoints, and learns if additional search is still likely to improve the current best candidate. Later, stopping is decided by comparing predicted continuation need with a cost measured in TEPs. Experimental results across multiple code families show that the proposed protocol significantly reduces average TEP count with only marginal FER degradation, using a single global model for the range of all operating signal-to-noise ratios (SNRs).