Polarized Element-pair Code Based FFMA over a Gaussian Multiple-access Channel

TL;DR

This paper introduces polarized element-pair codes for polarization-adjusted finite-field multiple-access systems, improving multiuser error performance and capacity in Gaussian channels through novel code design, power allocation, and decoding algorithms.

Contribution

It proposes a new polarized EP code tailored for Gaussian multiple access channels, along with optimal power allocation and specialized decoding algorithms, advancing multiuser communication performance.

Findings

Achieves a 1.25 dB coding gain for 15 users compared to existing systems.

Derives the channel capacity for the proposed polarized EP code system.

Develops two decoding algorithms: SCL and TopL-BMD, with comparable error performance.

Abstract

This paper presents polarized element-pair (EP) codes for polarization-adjusted finite-field multiple-access (PA-FFMA) systems. The core innovation of FFMA systems lies in their unique processing order that exchanges the conventional sequence of channel coding and multiplexing operations, effectively solving the multiuser finite-blocklength (FBL) problem while enhancing error performance. In this architecture, EPs serve as virtual resources for user separation, where different EP codes provide distinct error performance characteristics. The proposed polarized EP code differs from classical polar codes in one aspect that it is specifically designed for Gaussian multiple access channel (GMAC) environments rather than single-user Gaussian channels. We derive the channel capacity for this polarized EP code based FFMA system, then develop an optimal power allocation scheme to maximize multiuser channel capacity. The code construction employs the Marto Loco method for selecting the polarized index set. For decoding, we introduce two specialized algorithms. A successive cancellation list (SCL) decoder for the balanced information-parity section scenarios, and a top $L$ bifurcated minimum distance (Top$L$-BMD) decoder for small payload cases while maintaining comparable error performance. Simulations show that, for $15$ users, our system achieves a $1.25$ dB coding gain compared to the state-of-the-art polar random spreading systems.