Ordered-Statistics Decoding with Adaptive Gaussian Elimination Reduction for Short Codes

TL;DR

This paper introduces an adaptive Gaussian elimination reduction technique in ordered-statistics decoding that significantly reduces complexity at high SNRs without sacrificing error correction performance.

Contribution

It presents a novel adaptive GE reduction method with two decoding conditions, improving efficiency of OSD for short codes.

Findings

Reduces decoding complexity at high SNRs

Maintains error-correction performance

Breaks the complexity floor of traditional OSD

Abstract

In this paper, we propose an efficient ordered-statistics decoding (OSD) algorithm with an adaptive Gaussian elimination (GE) reduction technique. The proposed decoder utilizes two decoding conditions to adaptively remove GE in OSD. The first condition determines whether GE could be skipped in the OSD process by estimating the decoding error probability. Then, the second condition is utilized to identify the correct decoding result during the decoding process without GE. The proposed decoder can break the ``complexity floor'' in OSD decoders introduced by the GE overhead. Simulation results advise that when compared with the latest schemes in the literature, the proposed approach can significantly reduce the decoding complexity at high SNRs without any degradation in the error-correction capability.