Cargando…

Hunting Network Anomalies in a Railway Axle Counter System

This paper presents a comprehensive investigation of machine learning-based intrusion detection methods to reveal cyber attacks in railway axle counting networks. In contrast to the state-of-the-art works, our experimental results are validated with testbed-based real-world axle counting components....

Descripción completa

Detalles Bibliográficos
Autores principales: Kuchar, Karel, Holasova, Eva, Pospisil, Ondrej, Ruotsalainen, Henri, Fujdiak, Radek, Wagner, Adrian
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10052167/
https://www.ncbi.nlm.nih.gov/pubmed/36991830
http://dx.doi.org/10.3390/s23063122
_version_ 1785015095460364288
author Kuchar, Karel
Holasova, Eva
Pospisil, Ondrej
Ruotsalainen, Henri
Fujdiak, Radek
Wagner, Adrian
author_facet Kuchar, Karel
Holasova, Eva
Pospisil, Ondrej
Ruotsalainen, Henri
Fujdiak, Radek
Wagner, Adrian
author_sort Kuchar, Karel
collection PubMed
description This paper presents a comprehensive investigation of machine learning-based intrusion detection methods to reveal cyber attacks in railway axle counting networks. In contrast to the state-of-the-art works, our experimental results are validated with testbed-based real-world axle counting components. Furthermore, we aimed to detect targeted attacks on axle counting systems, which have higher impacts than conventional network attacks. We present a comprehensive investigation of machine learning-based intrusion detection methods to reveal cyber attacks in railway axle counting networks. According to our findings, the proposed machine learning-based models were able to categorize six different network states (normal and under attack). The overall accuracy of the initial models was ca. 70–100% for the test data set in laboratory conditions. In operational conditions, the accuracy decreased to under 50%. To increase the accuracy, we introduce a novel input data-preprocessing method with the denoted gamma parameter. This increased the accuracy of the deep neural network model to 69.52% for six labels, 85.11% for five labels, and 92.02% for two labels. The gamma parameter also removed the dependence on the time series, enabled relevant classification of data in the real network, and increased the accuracy of the model in real operations. This parameter is influenced by simulated attacks and, thus, allows the classification of traffic into specified classes.
format Online
Article
Text
id pubmed-10052167
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-100521672023-03-30 Hunting Network Anomalies in a Railway Axle Counter System Kuchar, Karel Holasova, Eva Pospisil, Ondrej Ruotsalainen, Henri Fujdiak, Radek Wagner, Adrian Sensors (Basel) Article This paper presents a comprehensive investigation of machine learning-based intrusion detection methods to reveal cyber attacks in railway axle counting networks. In contrast to the state-of-the-art works, our experimental results are validated with testbed-based real-world axle counting components. Furthermore, we aimed to detect targeted attacks on axle counting systems, which have higher impacts than conventional network attacks. We present a comprehensive investigation of machine learning-based intrusion detection methods to reveal cyber attacks in railway axle counting networks. According to our findings, the proposed machine learning-based models were able to categorize six different network states (normal and under attack). The overall accuracy of the initial models was ca. 70–100% for the test data set in laboratory conditions. In operational conditions, the accuracy decreased to under 50%. To increase the accuracy, we introduce a novel input data-preprocessing method with the denoted gamma parameter. This increased the accuracy of the deep neural network model to 69.52% for six labels, 85.11% for five labels, and 92.02% for two labels. The gamma parameter also removed the dependence on the time series, enabled relevant classification of data in the real network, and increased the accuracy of the model in real operations. This parameter is influenced by simulated attacks and, thus, allows the classification of traffic into specified classes. MDPI 2023-03-14 /pmc/articles/PMC10052167/ /pubmed/36991830 http://dx.doi.org/10.3390/s23063122 Text en © 2023 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Kuchar, Karel
Holasova, Eva
Pospisil, Ondrej
Ruotsalainen, Henri
Fujdiak, Radek
Wagner, Adrian
Hunting Network Anomalies in a Railway Axle Counter System
title Hunting Network Anomalies in a Railway Axle Counter System
title_full Hunting Network Anomalies in a Railway Axle Counter System
title_fullStr Hunting Network Anomalies in a Railway Axle Counter System
title_full_unstemmed Hunting Network Anomalies in a Railway Axle Counter System
title_short Hunting Network Anomalies in a Railway Axle Counter System
title_sort hunting network anomalies in a railway axle counter system
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10052167/
https://www.ncbi.nlm.nih.gov/pubmed/36991830
http://dx.doi.org/10.3390/s23063122
work_keys_str_mv AT kucharkarel huntingnetworkanomaliesinarailwayaxlecountersystem
AT holasovaeva huntingnetworkanomaliesinarailwayaxlecountersystem
AT pospisilondrej huntingnetworkanomaliesinarailwayaxlecountersystem
AT ruotsalainenhenri huntingnetworkanomaliesinarailwayaxlecountersystem
AT fujdiakradek huntingnetworkanomaliesinarailwayaxlecountersystem
AT wagneradrian huntingnetworkanomaliesinarailwayaxlecountersystem