Abstract

Wireless Sensor Networks (WSNs) are vulnerable to faults because of their deployment in unpredictable and hazardous environments. This makes WSN prone to failures such as software, hardware, and communication failures. Due to the sensor’s limited resources and diverse deployment fields, fault detection in WSNs has become a daunting task. To solve this problem, Support Vector Machine (SVM), Convolutional Neural Network (CNN), Stochastic Gradient Descent (SGD), Multilayer Perceptron (MLP), Random Forest (RF), and Probabilistic Neural Network (PNN) classifiers are used for classification of gain, offset, spike, data loss, out of bounds, and stuck-at faults at the sensor level. Out of six faults, two of them are induced in the datasets, i.e., spike and data loss faults. The results are compared on the basis of their Detection Accuracy (DA), True Positive Rate (TPR), Matthews Correlation Coefficients (MCC), and F1-score. In this paper, a comparative analysis is performed among the classifiers mentioned previously on real-world datasets. Simulations show that the RF algorithm secures a better fault detection rate than the rest of the classifiers.

Highlights

  • Wireless Sensor Networks (WSNs) are a set of independent sensor devices, which are connected by wireless channels

  • Each vector was composed of data collected at three successive instances t0, t1, t2, and each instance was constructed from two temperature measurements and two humidity measurements T1, T2 and H1, H2

  • The datasets consist of observation vectors Vt, which are composed of two humidity, i.e., H1, H2, and two temperature measurements, i.e., T1, T2, at three successive instances

Read more

Summary

Introduction

Wireless Sensor Networks (WSNs) are a set of independent sensor devices, which are connected by wireless channels These are comprehending structures that deal with the surroundings closely. Some techniques are distributed, centralized, or hybrid [3] They are mostly based on dynamics, self detection, and machine learning. Classification is one of the renowned approaches of data mining, which is a subset of machine learning It separates data into different classes explicitly and helps in decision making [4]. For the purpose of fault detection, Support Vector Machine (SVM), Convolutional Neural Network (CNN), Multilayer Perceptron (MLP), Stochastic Gradient Descent (SGD), Random Forest (RF), and Probabilistic Neural Network (PNN) classifiers, as shown, are used to classify the sensed data into two cases, i.e., normal or abnormal case For the purpose of fault detection, Support Vector Machine (SVM), Convolutional Neural Network (CNN), Multilayer Perceptron (MLP), Stochastic Gradient Descent (SGD), Random Forest (RF), and Probabilistic Neural Network (PNN) classifiers, as shown in Figure 1, are used to classify the sensed data into two cases, i.e., normal or abnormal case

Motivation
Challenges and Problem Statement
Contributions
Related Work
Future Work
Faults in WSNs
Gain Fault
Offset Fault
Stuck-at Fault
Spike Fault
Data Loss Fault
Out of Bounds
Classifiers
Phase 1
Phase 2
Phase 3
Simulations and Results
Datasets
Results
Conclusions and Future Work
Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call