Pre-configured Error Pattern Ordered Statistics Decoding for CRC-Polar Codes

TL;DR

This paper introduces PEPOSD, a novel decoding algorithm for CRC-polar codes that tests error patterns efficiently, improving accuracy and complexity trade-offs over existing methods.

Contribution

The paper proposes a pre-configured error pattern OSD algorithm for CRC-polar codes, inspired by GRAND, with optimized search orders and an offline-online structure for faster decoding.

Findings

PEPOSD outperforms CRC-aided OSD and list decoding in accuracy and complexity.

The offline-online structure accelerates the decoding process.

Optimized search orders improve error pattern testing efficiency.

Abstract

In this paper, we propose a pre-configured error pattern ordered statistics decoding (PEPOSD) algorithm and discuss its application to short cyclic redundancy check (CRC)-polar codes. Unlike the traditional OSD that changes the most reliable independent symbols, we regard the decoding process as testing the error patterns, like guessing random additive noise decoding (GRAND). Also, the pre-configurator referred from ordered reliability bits (ORB) GRAND can better control the range and testing order of EPs. Offline-online structure can accelerate the decoding process. Additionally, we also introduce two orders to optimize the search order for testing EPs. Compared with CRC-aided OSD and list decoding, PEPOSD can achieve a better trade-off between accuracy and complexity.