Constellation Design for Deep Joint Source-Channel Coding

TL;DR

This paper proposes two methods to map continuous deep joint source-channel coding (JSCC) outputs to finite constellations, simplifying RF implementation while maintaining performance, and demonstrates their effectiveness over traditional methods.

Contribution

It introduces two novel constellation mapping techniques for deep JSCC that improve practicality with minimal additional complexity.

Findings

Proposed methods outperform traditional QAM mapping.

Methods work effectively across various SNR levels.

Minimal increase in parameters needed for better performance.

Abstract

Deep learning-based joint source-channel coding (JSCC) has shown excellent performance in image and feature transmission. However, the output values of the JSCC encoder are continuous, which makes the constellation of modulation complex and dense. It is hard and expensive to design radio frequency chains for transmitting such full-resolution constellation points. In this paper, two methods of mapping the full-resolution constellation to finite constellation are proposed for real system implementation. The constellation mapping results of the proposed methods correspond to regular constellation and irregular constellation, respectively. We apply the methods to existing deep JSCC models and evaluate them on AWGN channels with different signal-to-noise ratios (SNRs). Experimental results show that the proposed methods outperform the traditional uniform quadrature amplitude modulation (QAM) constellation mapping method by only adding a few additional parameters.