Two-Class Joint Source-Channel Coding: Expurgated Exponents with i.i.d. Distributions

TL;DR

This paper investigates expurgated exponents in joint source-channel coding with i.i.d. distributions, demonstrating that a two-class source partitioning can outperform single-class coding in terms of error exponents.

Contribution

It introduces a two-class partitioning scheme for source sequences that improves error exponents over traditional single-class methods in joint source-channel coding.

Findings

Two-class partitioning achieves higher exponents than single-class coding.

Codeword distribution depending on source type enhances performance.

The method applies to discrete memoryless sources and channels.

Abstract

This paper studies expurgated exponents for joint source-channel coding of discrete memoryless sources and channels under i.i.d. random coding. We show that a two-class partitioning of source sequences, where the codeword distribution depends on the source type, achieves an exponent at least as high as that of optimal single-class coding, in which the codeword distribution is independent of the source message.