Neural Network-based Information-Theoretic Transceivers for High-Order Modulation Schemes

TL;DR

This paper introduces neural network-based transceivers for high-order modulation, demonstrating improved efficiency and performance through joint optimization and autoencoder architectures, with analysis of BER and SNR effects.

Contribution

It presents a novel NN-based bitwise receiver and an autoencoder-based end-to-end system optimized for high-order modulation schemes, enhancing efficiency and performance.

Findings

NN-based receiver maintains performance with improved efficiency

AE-based system outperforms baseline architectures

Training SNR significantly impacts system performance

Abstract

Neural network (NN)-based end-to-end (E2E) communication systems, in which each system component may consist of a portion of a neural network, have been investigated as potential tools for developing artificial intelligence (Al)-native E2E systems. In this paper, we propose an NN-based bitwise receiver that improves computational efficiency while maintaining performance comparable to baseline demappers. Building on this foundation, we introduce a novel symbol-wise autoencoder (AE)-based E2E system that jointly optimizes the transmitter and receiver at the physical layer. We evaluate the proposed NN-based receiver using bit-error rate (BER) analysis to confirm that the numerical BER achieved by NN-based receivers or transceivers is accurate. Results demonstrate that the AE-based system outperforms baseline architectures, particularly for higher-order modulation schemes. We further show that the training signal-to-noise ratio (SNR) significantly affects the performance of the systems when inference is conducted at different SNR levels.