Faults in Sensory Readings

Classification and Model Learning

Valentina Baljak, Kenji Tei, Shinichi Honiden

Research output: Contribution to specialist publicationArticle

2 Citations (Scopus)

Abstract

Faults in wireless sensor networks are a common occurrence and their accumulation can have a significant negative influence on the reliability of the network. Accuracy of sensory readings decreases over time. We focus on detection of faults as they can be observed in sensory readings. Trace that faults leave in data can be used for classification of faults independently of the underlying cause. We propose a complete and consistent fault classification based on two aspects. The first aspect is continuity and frequency of the occurrence, and the second is the existence of observable and learnable patterns. The network can learn a model of a fault from the past behavior and patterns visible in the data. We rely on centralized and straightforward detection methods using neighborhood vote and time series analysis. For the full classification phase, we propose the use of statistical pattern recognition, specifically, decision trees and regression. Current results show that this method works comparatively well when applied to dense data-centric wireless sensor network.

Original languageEnglish
Pages177-187
Number of pages11
Volume18
No.SPEC.ISS.1
Specialist publicationSensors and Transducers
Publication statusPublished - 2013 Jul 23
Externally publishedYes

Fingerprint

Wireless sensor networks
Time series analysis
Decision trees
Pattern recognition

Keywords

  • Fault tolerance
  • Model learning
  • Statistical pattern recognition
  • Wireless sensor networks

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Electrical and Electronic Engineering

Cite this

Faults in Sensory Readings : Classification and Model Learning. / Baljak, Valentina; Tei, Kenji; Honiden, Shinichi.

In: Sensors and Transducers, Vol. 18, No. SPEC.ISS.1, 23.07.2013, p. 177-187.

Research output: Contribution to specialist publicationArticle

@misc{e697126b642c4bbe90034a2fa5623660,
title = "Faults in Sensory Readings: Classification and Model Learning",
abstract = "Faults in wireless sensor networks are a common occurrence and their accumulation can have a significant negative influence on the reliability of the network. Accuracy of sensory readings decreases over time. We focus on detection of faults as they can be observed in sensory readings. Trace that faults leave in data can be used for classification of faults independently of the underlying cause. We propose a complete and consistent fault classification based on two aspects. The first aspect is continuity and frequency of the occurrence, and the second is the existence of observable and learnable patterns. The network can learn a model of a fault from the past behavior and patterns visible in the data. We rely on centralized and straightforward detection methods using neighborhood vote and time series analysis. For the full classification phase, we propose the use of statistical pattern recognition, specifically, decision trees and regression. Current results show that this method works comparatively well when applied to dense data-centric wireless sensor network.",
keywords = "Fault tolerance, Model learning, Statistical pattern recognition, Wireless sensor networks",
author = "Valentina Baljak and Kenji Tei and Shinichi Honiden",
year = "2013",
month = "7",
day = "23",
language = "English",
volume = "18",
pages = "177--187",
journal = "Sensors and Transducers",
issn = "1726-5479",
publisher = "International Frequency Sensor Association (IFSA)",

}

TY - GEN

T1 - Faults in Sensory Readings

T2 - Classification and Model Learning

AU - Baljak, Valentina

AU - Tei, Kenji

AU - Honiden, Shinichi

PY - 2013/7/23

Y1 - 2013/7/23

N2 - Faults in wireless sensor networks are a common occurrence and their accumulation can have a significant negative influence on the reliability of the network. Accuracy of sensory readings decreases over time. We focus on detection of faults as they can be observed in sensory readings. Trace that faults leave in data can be used for classification of faults independently of the underlying cause. We propose a complete and consistent fault classification based on two aspects. The first aspect is continuity and frequency of the occurrence, and the second is the existence of observable and learnable patterns. The network can learn a model of a fault from the past behavior and patterns visible in the data. We rely on centralized and straightforward detection methods using neighborhood vote and time series analysis. For the full classification phase, we propose the use of statistical pattern recognition, specifically, decision trees and regression. Current results show that this method works comparatively well when applied to dense data-centric wireless sensor network.

AB - Faults in wireless sensor networks are a common occurrence and their accumulation can have a significant negative influence on the reliability of the network. Accuracy of sensory readings decreases over time. We focus on detection of faults as they can be observed in sensory readings. Trace that faults leave in data can be used for classification of faults independently of the underlying cause. We propose a complete and consistent fault classification based on two aspects. The first aspect is continuity and frequency of the occurrence, and the second is the existence of observable and learnable patterns. The network can learn a model of a fault from the past behavior and patterns visible in the data. We rely on centralized and straightforward detection methods using neighborhood vote and time series analysis. For the full classification phase, we propose the use of statistical pattern recognition, specifically, decision trees and regression. Current results show that this method works comparatively well when applied to dense data-centric wireless sensor network.

KW - Fault tolerance

KW - Model learning

KW - Statistical pattern recognition

KW - Wireless sensor networks

UR - http://www.scopus.com/inward/record.url?scp=84880270057&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84880270057&partnerID=8YFLogxK

M3 - Article

VL - 18

SP - 177

EP - 187

JO - Sensors and Transducers

JF - Sensors and Transducers

SN - 1726-5479

ER -