Fault detection in wireless sensor networks

A machine learning approach

Ehsan Ullah Warriach, Kenji Tei

Research output: Contribution to conferencePaper

15 Citations (Scopus)

Abstract

Wireless Sensor Network (WSN) deployment experiences show that collected data is prone to be faulty. Faults are due to internal and external influences, such as calibration, low battery, environmental interference and sensor aging. However, only few solutions exist to deal with faulty sensory data in WSN. We develop a statistical approach to detect and identify faults in a WSN. In particular, we focus on the identification and classification of data and system fault types as it is essential to perform accurate recovery actions. Our method uses Hidden Markov Models (HMMs) to capture the fault-free dynamics of an environment and dynamics of faulty data. It then performs a structural analysis of these HMMs to determine the type of data and system faults affecting sensor measurements. The approach is validated using real data obtained from over one month of samples from motes deployed in an actual living lab.

Original languageEnglish
Pages758-765
Number of pages8
DOIs
Publication statusPublished - 2013 Dec 1
Externally publishedYes
Event2013 16th IEEE International Conference on Computational Science and Engineering, CSE 2013 - Sydney, NSW
Duration: 2013 Dec 32013 Dec 5

Other

Other2013 16th IEEE International Conference on Computational Science and Engineering, CSE 2013
CitySydney, NSW
Period13/12/313/12/5

Fingerprint

Fault detection
Learning systems
Wireless sensor networks
Hidden Markov models
Sensors
Structural analysis
Aging of materials
Calibration
Recovery

ASJC Scopus subject areas

  • Computer Science (miscellaneous)

Cite this

Warriach, E. U., & Tei, K. (2013). Fault detection in wireless sensor networks: A machine learning approach. 758-765. Paper presented at 2013 16th IEEE International Conference on Computational Science and Engineering, CSE 2013, Sydney, NSW, . https://doi.org/10.1109/CSE.2013.116

Fault detection in wireless sensor networks : A machine learning approach. / Warriach, Ehsan Ullah; Tei, Kenji.

2013. 758-765 Paper presented at 2013 16th IEEE International Conference on Computational Science and Engineering, CSE 2013, Sydney, NSW, .

Research output: Contribution to conferencePaper

Warriach, EU & Tei, K 2013, 'Fault detection in wireless sensor networks: A machine learning approach' Paper presented at 2013 16th IEEE International Conference on Computational Science and Engineering, CSE 2013, Sydney, NSW, 13/12/3 - 13/12/5, pp. 758-765. https://doi.org/10.1109/CSE.2013.116
Warriach EU, Tei K. Fault detection in wireless sensor networks: A machine learning approach. 2013. Paper presented at 2013 16th IEEE International Conference on Computational Science and Engineering, CSE 2013, Sydney, NSW, . https://doi.org/10.1109/CSE.2013.116
Warriach, Ehsan Ullah ; Tei, Kenji. / Fault detection in wireless sensor networks : A machine learning approach. Paper presented at 2013 16th IEEE International Conference on Computational Science and Engineering, CSE 2013, Sydney, NSW, .8 p.
@conference{5204c508d97c43d390746551c48b1d0d,
title = "Fault detection in wireless sensor networks: A machine learning approach",
abstract = "Wireless Sensor Network (WSN) deployment experiences show that collected data is prone to be faulty. Faults are due to internal and external influences, such as calibration, low battery, environmental interference and sensor aging. However, only few solutions exist to deal with faulty sensory data in WSN. We develop a statistical approach to detect and identify faults in a WSN. In particular, we focus on the identification and classification of data and system fault types as it is essential to perform accurate recovery actions. Our method uses Hidden Markov Models (HMMs) to capture the fault-free dynamics of an environment and dynamics of faulty data. It then performs a structural analysis of these HMMs to determine the type of data and system faults affecting sensor measurements. The approach is validated using real data obtained from over one month of samples from motes deployed in an actual living lab.",
author = "Warriach, {Ehsan Ullah} and Kenji Tei",
year = "2013",
month = "12",
day = "1",
doi = "10.1109/CSE.2013.116",
language = "English",
pages = "758--765",
note = "2013 16th IEEE International Conference on Computational Science and Engineering, CSE 2013 ; Conference date: 03-12-2013 Through 05-12-2013",

}

TY - CONF

T1 - Fault detection in wireless sensor networks

T2 - A machine learning approach

AU - Warriach, Ehsan Ullah

AU - Tei, Kenji

PY - 2013/12/1

Y1 - 2013/12/1

N2 - Wireless Sensor Network (WSN) deployment experiences show that collected data is prone to be faulty. Faults are due to internal and external influences, such as calibration, low battery, environmental interference and sensor aging. However, only few solutions exist to deal with faulty sensory data in WSN. We develop a statistical approach to detect and identify faults in a WSN. In particular, we focus on the identification and classification of data and system fault types as it is essential to perform accurate recovery actions. Our method uses Hidden Markov Models (HMMs) to capture the fault-free dynamics of an environment and dynamics of faulty data. It then performs a structural analysis of these HMMs to determine the type of data and system faults affecting sensor measurements. The approach is validated using real data obtained from over one month of samples from motes deployed in an actual living lab.

AB - Wireless Sensor Network (WSN) deployment experiences show that collected data is prone to be faulty. Faults are due to internal and external influences, such as calibration, low battery, environmental interference and sensor aging. However, only few solutions exist to deal with faulty sensory data in WSN. We develop a statistical approach to detect and identify faults in a WSN. In particular, we focus on the identification and classification of data and system fault types as it is essential to perform accurate recovery actions. Our method uses Hidden Markov Models (HMMs) to capture the fault-free dynamics of an environment and dynamics of faulty data. It then performs a structural analysis of these HMMs to determine the type of data and system faults affecting sensor measurements. The approach is validated using real data obtained from over one month of samples from motes deployed in an actual living lab.

UR - http://www.scopus.com/inward/record.url?scp=84900366089&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84900366089&partnerID=8YFLogxK

U2 - 10.1109/CSE.2013.116

DO - 10.1109/CSE.2013.116

M3 - Paper

SP - 758

EP - 765

ER -