On Systematic Polarization-Adjusted Convolutional (PAC) Codes

TL;DR

This paper introduces systematic PAC codes with improved bit-error rates and presents a genetic-algorithm based construction method that enables these codes to approach theoretical bounds for larger code sizes.

Contribution

The paper proposes systematic PAC codes and a genetic-algorithm based construction method to enhance performance and approach theoretical bounds for larger code sizes.

Findings

Systematic PAC codes improve bit-error rate by up to 0.2 dB.

The proposed construction method approaches theoretical bounds for (256,128) codes.

Systematic and non-systematic PAC codes perform similarly in approaching bounds.

Abstract

Polarization-adjusted convolutional (PAC) codes were recently proposed and arouse the interest of the channel coding community because they were shown to approach theoretical bounds for the (128,64) code size. In this letter, we propose systematic PAC codes. Thanks to the systematic property, improvement in the bit-error rate of up to 0.2 dB is observed, while preserving the frame-error rate performance. Moreover, a genetic-algorithm based construction method targeted to approach the theoretical bound is provided. It is then shown that using the proposed construction method systematic and non-systematic PAC codes can approach the theoretical bound even for higher code sizes such as (256,128).