Improved Random-Binning Exponent for Distributed Hypothesis Testing

TL;DR

This paper improves the error exponent in distributed binary hypothesis testing by modifying the decision rule in an existing quantization and binning scheme, leading to better performance.

Contribution

It introduces a simple modification to the receiver's decision rule that achieves a higher error exponent in distributed hypothesis testing.

Findings

Achieves a better error exponent than previous methods

Simplifies the decision rule in the existing scheme

Provides theoretical analysis of the improved exponent

Abstract

Consider the problem of distributed binary hypothesis testing with two terminals, where the decision is made at one of them (the "receiver"). We study the exponent of the error probability of the second type. Previously, an achievable exponent was derived by Shimokawa, Han, and Amari using a "quantization and binning" scheme. We propose a simple modification on the receiver's decision rule in this scheme to attain a better exponent.