Distributed Change Detection via Average Consensus over Networks

TL;DR

This paper introduces a distributed change detection algorithm using average consensus in sensor networks, which matches centralized performance and effectively detects asynchronous changes in real-time.

Contribution

It proposes a novel distributed detection method based on local CUSUM statistics and consensus, with theoretical performance guarantees and practical effectiveness.

Findings

Performance bounds match centralized algorithms under mild conditions.

Effective detection of asynchronous changes demonstrated.

Algorithm performs well in numerical experiments.

Abstract

Distributed change-point detection has been a fundamental problem when performing real-time monitoring using sensor-networks. We propose a distributed detection algorithm, where each sensor only exchanges CUSUM statistic with their neighbors based on the average consensus scheme, and an alarm is raised when local consensus statistic exceeds a pre-specified global threshold. We provide theoretical performance bounds showing that the performance of the fully distributed scheme can match the centralized algorithms under some mild conditions. Numerical experiments demonstrate the good performance of the algorithm especially in detecting asynchronous changes.