Hypothesis Testing and Decision Theoretic Approach for Fault Detection in Wireless Sensor Networks

TL;DR

This paper presents a decision-theoretic framework for fault detection in wireless sensor networks, addressing both measurement noise and sensor faults using Neyman-Pearson and Bayes tests.

Contribution

It introduces novel fault detection schemes that incorporate error probabilities into the optimal event detection process in WSNs.

Findings

Fault detection schemes effectively identify sensor faults and noise.

The proposed methods improve detection accuracy in WSNs.

The approach integrates statistical hypothesis testing into fault detection.

Abstract

Sensor networks aim at monitoring their surroundings for event detection and object tracking. But due to failure or death of sensors, false signal can be transmitted. In this paper, we consider the problem of fault detection in wireless sensor network (WSN), in particular, addressing both the noise-related measurement error and sensor fault simultaneously in fault detection. We assume that the sensors are placed at the center of a square (or hexagonal) cell in region of interest (ROI) and, if the event occurs, it occurs at a particular cell of the ROI. We propose fault detection schemes that take into account error probabilities into the optimal event detection process. We develop the schemes under the consideration of Neyman-Pearson test and Bayes test.