Anomaly Detection in Particulate Matter Sensor using Hypothesis Pruning Generative Adversarial Network

TL;DR

This paper introduces HP-GAN, a novel anomaly detection method using hypothesis pruning in generative adversarial networks, to identify malfunctions in cost-effective PM sensors, improving maintenance accuracy.

Contribution

The paper presents a new HP-GAN architecture specifically designed for anomaly detection in TEOM-based PM sensors, outperforming existing methods.

Findings

HP-GAN achieves higher detection accuracy than baseline models.

Experimental results validate the effectiveness of hypothesis pruning in GANs.

The approach enhances sensor maintenance and reliability.

Abstract

World Health Organization (WHO) provides the guideline for managing the Particulate Matter (PM) level because when the PM level is higher, it threats the human health. For managing PM level, the procedure for measuring PM value is needed firstly. We use Tapered Element Oscillating Microbalance (TEOM)-based PM measuring sensors because it shows higher cost-effectiveness than Beta Attenuation Monitor (BAM)-based sensor. However, TEOM-based sensor has higher probability of malfunctioning than BAM-based sensor. In this paper, we call the overall malfunction as an anomaly, and we aim to detect anomalies for the maintenance of PM measuring sensors. We propose a novel architecture for solving the above aim that named as Hypothesis Pruning Generative Adversarial Network (HP-GAN). We experimentally compare the several anomaly detection architectures to certify ours performing better.