A machine learning technique in a multi-agent framework for online outliers detection in Wireless Sensor Networks

H. Martins; F. Januário; L. Palma; A. Cardoso; Paulo José Carrilho de Sousa Gil

doi:10.1109/IECON.2015.7392180

A machine learning technique in a multi-agent framework for online outliers detection in Wireless Sensor Networks

H. Martins, F. Januário, L. Palma, A. Cardoso, Paulo José Carrilho de Sousa Gil

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

18 Citations (Scopus)

Abstract

Wireless Sensor Networks enable flexibility, low operational and maintenance costs, as well as scalability in a variety of scenarios. However, in the context of industrial monitoring scenarios the use of Wireless Sensor Networks can compromise the system's performance due to several factors, being one of them the presence of outliers in raw data. In order to improve the overall system's resilience, this paper proposes a distributed hierarchical multi-agent architecture where each agent is responsible for a specific task. This paper deals with online detection and accommodation of outliers in non-stationary time-series by appealing to a machine learning technique. The methodology is based on a Least Squares Support Vector Machine along with a sliding window-based learning algorithm. A modification to this method is considered to improve its performance in transient raw data collected from transmitters over a Wireless Sensor Networks (WSNs). An empirical study based on laboratory test-bed show the feasibility and relevance of incorporating the proposed methodology in the context of monitoring systems over Wireless Sensor Networks.

Original language	English
Title of host publication	Industrial Electronics Society, IECON 2015 - 41st Annual Conference of the IEEE
Pages	688-693
Number of pages	6
DOIs	https://doi.org/10.1109/IECON.2015.7392180
Publication status	Published - 1 Nov 2015
Event	41st Annual Conference of the IEEE Industrial Electronics Society, IECON 2015 - Yokohama, Japan Duration: 9 Nov 2015 → 12 Nov 2015

Conference

Conference	41st Annual Conference of the IEEE Industrial Electronics Society, IECON 2015
Country/Territory	Japan
City	Yokohama
Period	9/11/15 → 12/11/15

Keywords

Context, data acquisition, distributed hierarchical multiagent architecture, industrial monitoring, Kernel, learning (artificial intelligence), least squares approximations, least square support vector machine, machine learning technique, Memory, monitoring, multiagent framework, Multi-agent systems, nonstationary time-series, online outlier detection, sliding window-based learning algorithm, support vector machines, Symmetric matrices, time series, transmitter, transmitters, wireless sensor network, wireless sensor networks, WSN
Automation & Control Systems
Engineering, Electrical & Electronic

Access to Document

10.1109/IECON.2015.7392180Licence: Unspecified

Cite this

@inproceedings{1ba21ccdbd374e46a91928d290972346,

title = "A machine learning technique in a multi-agent framework for online outliers detection in Wireless Sensor Networks",

abstract = "Wireless Sensor Networks enable flexibility, low operational and maintenance costs, as well as scalability in a variety of scenarios. However, in the context of industrial monitoring scenarios the use of Wireless Sensor Networks can compromise the system's performance due to several factors, being one of them the presence of outliers in raw data. In order to improve the overall system's resilience, this paper proposes a distributed hierarchical multi-agent architecture where each agent is responsible for a specific task. This paper deals with online detection and accommodation of outliers in non-stationary time-series by appealing to a machine learning technique. The methodology is based on a Least Squares Support Vector Machine along with a sliding window-based learning algorithm. A modification to this method is considered to improve its performance in transient raw data collected from transmitters over a Wireless Sensor Networks (WSNs). An empirical study based on laboratory test-bed show the feasibility and relevance of incorporating the proposed methodology in the context of monitoring systems over Wireless Sensor Networks.",

keywords = "Context, data acquisition, distributed hierarchical multiagent architecture, industrial monitoring, Kernel, learning (artificial intelligence), least squares approximations, least square support vector machine, machine learning technique, Memory, monitoring, multiagent framework, Multi-agent systems, nonstationary time-series, online outlier detection, sliding window-based learning algorithm, support vector machines, Symmetric matrices, time series, transmitter, transmitters, wireless sensor network, wireless sensor networks, WSN, Automation & Control Systems, Engineering, Electrical & Electronic",

author = "H. Martins and F. Janu{\'a}rio and L. Palma and A. Cardoso and Gil, {Paulo Jos{\'e} Carrilho de Sousa}",

note = "sem pdf conforme despacho.; 41st Annual Conference of the IEEE Industrial Electronics Society, IECON 2015 ; Conference date: 09-11-2015 Through 12-11-2015",

year = "2015",

month = nov,

day = "1",

doi = "10.1109/IECON.2015.7392180",

language = "English",

pages = "688--693",

booktitle = "Industrial Electronics Society, IECON 2015 - 41st Annual Conference of the IEEE",

}

Martins, H, Januário, F, Palma, L, Cardoso, A & Gil, PJCDS 2015, A machine learning technique in a multi-agent framework for online outliers detection in Wireless Sensor Networks. in Industrial Electronics Society, IECON 2015 - 41st Annual Conference of the IEEE. pp. 688-693, 41st Annual Conference of the IEEE Industrial Electronics Society, IECON 2015, Yokohama, Japan, 9/11/15. https://doi.org/10.1109/IECON.2015.7392180

TY - GEN

T1 - A machine learning technique in a multi-agent framework for online outliers detection in Wireless Sensor Networks

AU - Martins, H.

AU - Januário, F.

AU - Palma, L.

AU - Cardoso, A.

AU - Gil, Paulo José Carrilho de Sousa

N1 - sem pdf conforme despacho.

PY - 2015/11/1

Y1 - 2015/11/1

N2 - Wireless Sensor Networks enable flexibility, low operational and maintenance costs, as well as scalability in a variety of scenarios. However, in the context of industrial monitoring scenarios the use of Wireless Sensor Networks can compromise the system's performance due to several factors, being one of them the presence of outliers in raw data. In order to improve the overall system's resilience, this paper proposes a distributed hierarchical multi-agent architecture where each agent is responsible for a specific task. This paper deals with online detection and accommodation of outliers in non-stationary time-series by appealing to a machine learning technique. The methodology is based on a Least Squares Support Vector Machine along with a sliding window-based learning algorithm. A modification to this method is considered to improve its performance in transient raw data collected from transmitters over a Wireless Sensor Networks (WSNs). An empirical study based on laboratory test-bed show the feasibility and relevance of incorporating the proposed methodology in the context of monitoring systems over Wireless Sensor Networks.

AB - Wireless Sensor Networks enable flexibility, low operational and maintenance costs, as well as scalability in a variety of scenarios. However, in the context of industrial monitoring scenarios the use of Wireless Sensor Networks can compromise the system's performance due to several factors, being one of them the presence of outliers in raw data. In order to improve the overall system's resilience, this paper proposes a distributed hierarchical multi-agent architecture where each agent is responsible for a specific task. This paper deals with online detection and accommodation of outliers in non-stationary time-series by appealing to a machine learning technique. The methodology is based on a Least Squares Support Vector Machine along with a sliding window-based learning algorithm. A modification to this method is considered to improve its performance in transient raw data collected from transmitters over a Wireless Sensor Networks (WSNs). An empirical study based on laboratory test-bed show the feasibility and relevance of incorporating the proposed methodology in the context of monitoring systems over Wireless Sensor Networks.

KW - Context, data acquisition, distributed hierarchical multiagent architecture, industrial monitoring, Kernel, learning (artificial intelligence), least squares approximations, least square support vector machine, machine learning technique, Memory, monitori

KW - Automation & Control Systems

KW - Engineering, Electrical & Electronic

U2 - 10.1109/IECON.2015.7392180

DO - 10.1109/IECON.2015.7392180

M3 - Conference contribution

SP - 688

EP - 693

BT - Industrial Electronics Society, IECON 2015 - 41st Annual Conference of the IEEE

T2 - 41st Annual Conference of the IEEE Industrial Electronics Society, IECON 2015

Y2 - 9 November 2015 through 12 November 2015

ER -

A machine learning technique in a multi-agent framework for online outliers detection in Wireless Sensor Networks

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this