A New Inequality Related to Proofs of Strong Converse Theorems for Source or Channel Networks

TL;DR

This paper introduces a new inequality that advances the proof techniques for strong converse theorems in multiterminal information theory, leading to improved bounds for source coding problems.

Contribution

It presents a novel inequality and applies it to derive a tighter strong converse outer bound for the Wyner-Ziv source coding problem.

Findings

Derived a new strong converse outer bound for Wyner-Ziv coding

The outer bound deviates from the rate distortion region by O(1/√n)

Enhanced proof techniques for multiterminal information theory

Abstract

In this paper we provide a new inequality useful for the proofs of strong converse theorems in the multiterminal information theory. We apply this inequality to the recent work by Tyagi and Watanabe on the strong converse theorem for the Wyner-Ziv source coding problem to obtain a new strong converse outer bound. This outer bound deviates from the Wyner-Ziv rate distortion region with the order $O\left(\frac{1}{\sqrt{n}}\right)$ on the length $n$ of source outputs.