Strong Converse Exponent for Remote Lossy Source Coding

TL;DR

This paper investigates the exponential rate at which the probability of exceeding a distortion threshold approaches certainty in remote lossy source coding, providing bounds that unify and extend previous results.

Contribution

It establishes the strong converse exponent for remote lossy source coding, offering matched bounds and connecting to prior findings in related areas.

Findings

Derived the strong converse exponent with matching bounds

Unified previous results on lossy source coding and biometric authentication

Enhanced understanding of error probability convergence in source coding

Abstract

Past works on remote lossy source coding studied the rate under average distortion and the error exponent of excess distortion probability. In this work, we look into how fast the excess distortion probability converges to 1 at small rates, also known as exponential strong converse. We characterize its exponent by establishing matched upper and lower bounds. From the exponent, we also recover two previous results on lossy source coding and biometric authentication.