Rethinking IoT Network Reliability in the Era of Machine Learning

Xenofon Fafoutis, Letizia Marchegiani

Publikation: Bidrag til bog/antologi/rapport/konference proceedingKonferenceartikel i proceedingForskningpeer review

Resumé

In the Internet of Things (IoT), wireless sensor networks are often paired with machine learning frameworks to deliver applications of high societal impact and support critical infrastructures. In this context, this paper investigates the relationship between network reliability and the reliability of the machine learning framework in terms of prediction accuracy. Our experimental analysis leverages six data sets of various degrees of information redundancy and considers four machine learning algorithms that are commonly used for classification. In turn, packet loss is inserted in the raw input data, emulating various networking loss patterns in terms of burstiness. The experimental results consistently demonstrate a non-linear relationship between the reliability of the network and the accuracy of the machine learning classifier, indicating that not all data packets are equally valuable to the application performance. We conclude with recommendations for IoT practitioners and IoT system designers.

OriginalsprogEngelsk
Titel 2019 International Conference on Internet of Things (iThings) and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber, Physical and Social Computing (CPSCom) and IEEE Smart Data (SmartData)
Antal sider8
ForlagIEEE
Publikationsdato2019
Sider1112-1119
ISBN (Trykt)978-1-7281-2981-5
ISBN (Elektronisk)978-1-7281-2980-8
DOI
StatusUdgivet - 2019
Begivenhed2019 International Conference on Internet of Things (iThings) and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber, Physical and Social Computing (CPSCom) and IEEE Smart Data (SmartData) - Atlanta, USA
Varighed: 14 jul. 201917 jul. 2019

Konference

Konference2019 International Conference on Internet of Things (iThings) and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber, Physical and Social Computing (CPSCom) and IEEE Smart Data (SmartData)
LandUSA
ByAtlanta
Periode14/07/201917/07/2019

Fingerprint

Learning systems
Critical infrastructures
Packet loss
Learning algorithms
Redundancy
Wireless sensor networks
Classifiers
Internet of things

Citer dette

Fafoutis, X., & Marchegiani, L. (2019). Rethinking IoT Network Reliability in the Era of Machine Learning. I 2019 International Conference on Internet of Things (iThings) and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber, Physical and Social Computing (CPSCom) and IEEE Smart Data (SmartData) (s. 1112-1119). IEEE. https://doi.org/10.1109/iThings/GreenCom/CPSCom/SmartData.2019.00189
Fafoutis, Xenofon ; Marchegiani, Letizia. / Rethinking IoT Network Reliability in the Era of Machine Learning. 2019 International Conference on Internet of Things (iThings) and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber, Physical and Social Computing (CPSCom) and IEEE Smart Data (SmartData). IEEE, 2019. s. 1112-1119
@inproceedings{146e591c4f3942a0a6e0290008517182,
title = "Rethinking IoT Network Reliability in the Era of Machine Learning",
abstract = "In the Internet of Things (IoT), wireless sensor networks are often paired with machine learning frameworks to deliver applications of high societal impact and support critical infrastructures. In this context, this paper investigates the relationship between network reliability and the reliability of the machine learning framework in terms of prediction accuracy. Our experimental analysis leverages six data sets of various degrees of information redundancy and considers four machine learning algorithms that are commonly used for classification. In turn, packet loss is inserted in the raw input data, emulating various networking loss patterns in terms of burstiness. The experimental results consistently demonstrate a non-linear relationship between the reliability of the network and the accuracy of the machine learning classifier, indicating that not all data packets are equally valuable to the application performance. We conclude with recommendations for IoT practitioners and IoT system designers.",
author = "Xenofon Fafoutis and Letizia Marchegiani",
year = "2019",
doi = "10.1109/iThings/GreenCom/CPSCom/SmartData.2019.00189",
language = "English",
isbn = "978-1-7281-2981-5",
pages = "1112--1119",
booktitle = "2019 International Conference on Internet of Things (iThings) and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber, Physical and Social Computing (CPSCom) and IEEE Smart Data (SmartData)",
publisher = "IEEE",
address = "United States",

}

Fafoutis, X & Marchegiani, L 2019, Rethinking IoT Network Reliability in the Era of Machine Learning. i 2019 International Conference on Internet of Things (iThings) and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber, Physical and Social Computing (CPSCom) and IEEE Smart Data (SmartData). IEEE, s. 1112-1119, Atlanta, USA, 14/07/2019. https://doi.org/10.1109/iThings/GreenCom/CPSCom/SmartData.2019.00189

Rethinking IoT Network Reliability in the Era of Machine Learning. / Fafoutis, Xenofon; Marchegiani, Letizia.

2019 International Conference on Internet of Things (iThings) and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber, Physical and Social Computing (CPSCom) and IEEE Smart Data (SmartData). IEEE, 2019. s. 1112-1119.

Publikation: Bidrag til bog/antologi/rapport/konference proceedingKonferenceartikel i proceedingForskningpeer review

TY - GEN

T1 - Rethinking IoT Network Reliability in the Era of Machine Learning

AU - Fafoutis, Xenofon

AU - Marchegiani, Letizia

PY - 2019

Y1 - 2019

N2 - In the Internet of Things (IoT), wireless sensor networks are often paired with machine learning frameworks to deliver applications of high societal impact and support critical infrastructures. In this context, this paper investigates the relationship between network reliability and the reliability of the machine learning framework in terms of prediction accuracy. Our experimental analysis leverages six data sets of various degrees of information redundancy and considers four machine learning algorithms that are commonly used for classification. In turn, packet loss is inserted in the raw input data, emulating various networking loss patterns in terms of burstiness. The experimental results consistently demonstrate a non-linear relationship between the reliability of the network and the accuracy of the machine learning classifier, indicating that not all data packets are equally valuable to the application performance. We conclude with recommendations for IoT practitioners and IoT system designers.

AB - In the Internet of Things (IoT), wireless sensor networks are often paired with machine learning frameworks to deliver applications of high societal impact and support critical infrastructures. In this context, this paper investigates the relationship between network reliability and the reliability of the machine learning framework in terms of prediction accuracy. Our experimental analysis leverages six data sets of various degrees of information redundancy and considers four machine learning algorithms that are commonly used for classification. In turn, packet loss is inserted in the raw input data, emulating various networking loss patterns in terms of burstiness. The experimental results consistently demonstrate a non-linear relationship between the reliability of the network and the accuracy of the machine learning classifier, indicating that not all data packets are equally valuable to the application performance. We conclude with recommendations for IoT practitioners and IoT system designers.

U2 - 10.1109/iThings/GreenCom/CPSCom/SmartData.2019.00189

DO - 10.1109/iThings/GreenCom/CPSCom/SmartData.2019.00189

M3 - Article in proceeding

SN - 978-1-7281-2981-5

SP - 1112

EP - 1119

BT - 2019 International Conference on Internet of Things (iThings) and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber, Physical and Social Computing (CPSCom) and IEEE Smart Data (SmartData)

PB - IEEE

ER -

Fafoutis X, Marchegiani L. Rethinking IoT Network Reliability in the Era of Machine Learning. I 2019 International Conference on Internet of Things (iThings) and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber, Physical and Social Computing (CPSCom) and IEEE Smart Data (SmartData). IEEE. 2019. s. 1112-1119 https://doi.org/10.1109/iThings/GreenCom/CPSCom/SmartData.2019.00189