https://www.mdu.se/

mdu.sePublications
System disruptions
There are recurring service disruptions on some of the search portals due to high traffic. We are working to resolve the issue, you may temporarily encounter an error message.
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Empirical Evaluation of Deep Learning Models for Intrusion Detection in Industrial Control Networks
Mälardalen University, School of Innovation, Design and Engineering, Embedded Systems.ORCID iD: 0000-0001-5332-1033
RISE Research Institute of Sweden, Västerås, Sweden.
Cloud and Embedded Platform, Traton AB, Södertälje, Sweden.ORCID iD: 0000-0003-3354-1463
Mälardalen University, School of Innovation, Design and Engineering, Embedded Systems.ORCID iD: 0000-0002-7235-6888
(English)Manuscript (preprint) (Other academic)
Abstract [en]

Industrial Control Systems (ICS) are increasingly targeted by cyberattacks, raising the need for effective Intrusion Detection Systems (IDS). However, their operational environments impose strict constraints on the deployment of IDS, particularly regarding computational efficiency and resource usage. ML-based approaches have shown promising performance in providing intelligent cyber detection to detect unknown and evolving threats. In this study, we systematically compare several deep learning models for intrusion detection in ICS networks. Given the scarcity of attack-representing data, we focus on semi-supervised approaches that learn from normal behavior, including Autoencoder (AE), Variational Autoencoder (VAE), Conditional VAE (CVAE), Probabilistic VAE (PVAE), and LSTM Autoencoder (LSTM-AE). Motivated by the need for lightweight models, our evaluation emphasizes both detection performance and deployment-relevant metrics (e.g., RAM usage, CPU load, training/inference time, and model size). All tests were conducted on isolated, resource-constrained hardware that resembles typical ICS edge devices and was dedicated solely to the IDS workload. The results show that AE offers the best trade-off between accuracy and resource efficiency, achieving F1-scores above 0.94 with minimal computational footprint. Importantly, we found that model size alone does not determine resource usage or timing efficiency, as architectural complexity can lead smaller models to consume more resources and run slower than larger ones. Hyperparameter tuning also affected footprint differently across models, emphasizing the need to jointly consider architecture, tuning, and hardware constraints when selecting models for ICS deployment.

Keywords [en]
Industrial Control Systems (ICS), Network Intrusion Detection Systems (NIDS), Semi-supervised Deep Learning, Resource-Constrained, Model Efficiency
National Category
Embedded Systems Computer Systems Control Engineering
Research subject
Computer Science
Identifiers
URN: urn:nbn:se:mdh:diva-73129OAI: oai:DiVA.org:mdh-73129DiVA, id: diva2:1992882
Available from: 2025-08-28 Created: 2025-08-28 Last updated: 2025-10-10Bibliographically approved
In thesis
1. Machine Learning-Based Network Intrusion Detection for Industrial Control
Open this publication in new window or tab >>Machine Learning-Based Network Intrusion Detection for Industrial Control
2025 (English)Doctoral thesis, comprehensive summary (Other academic)
Abstract [en]

The growing connectivity of Industrial Control Systems (ICS) has increased their exposure to cyber threats, posing serious risks to both critical infrastructure (e.g., power plants, water systems) and industrial operations. Although Machine Learning (ML)-based Intrusion Detection Systems (IDS) show potential to detect complex and novel attacks, their practical application in ICS environments remains challenging. This thesis investigates the feasibility, design considerations, and barriers to applying ML-based IDS in ICSs. Key challenges include limited labeled data, high sensitivity to false alarms, distributed architectures, and constrained hardware. Through empirical evaluations and prototype implementations, we show how tailored ML and system design strategies can address these issues.

Our first goal is to enable experimentation with cybersecurity and intrusion detection in industrial systems. The thesis introduces ICSSIM, a novel framework for creating flexible, scalable, and cost-effective ICS testbeds.  It also presents ICSFlowGenerator, a tool for analyzing network traffic and computing customized network flow parameters. Using this tool, the thesis presents ICS-Flow, a new dataset developed to train and evaluate anomaly detection models, which serves as a realistic benchmark for assessing ML-based intrusion detection in ICS networks. 

In addition, the thesis tackles several technical deployment barriers. To support distributed ICS architectures and minimize the reliance on central servers, federated learning is explored as a decentralized training strategy. It also investigates semi-supervised learning techniques for detecting anomalies using only normal traffic.  To reduce the burden of false alarms, we integrate a decision support system with the IDS to filter alerts and suggest mitigation actions. Furthermore, the thesis emphasizes the importance of temporal traffic patterns in identifying attack types and evaluates the efficiency and resource demands of various ML models on ICS-representative hardware. Collectively, these contributions advance the practical application of ML-based intrusion detection in ICSs.

Abstract [sv]

Digitaliseringen och ökande uppkoppling av industriella styrsystem (ICS) har medfört ökad exponering för cyberhot och innebär allvarliga risker för både kritisk infrastruktur (t.ex. kraftverk, vattensystem) och industriella verksamheter. Även om intrångsdetekteringssystem (IDS) baserade på maskininlärning (ML) har potential att upptäcka komplexa och nya attacker, är deras praktiska tillämpning i ICS-miljöer fortfarande utmanande. I den här avhandlingen analyseras genomförbarheten, designöverväganden och hinder för att applicera ML-baserade IDS i ICS. Centrala utmaningar omfattar begränsad mängd annoterad data, hög känslighet för falsklarm, distribuerade arkitekturer och hårdvarubegränsningar. Genom empiriska utvärderingar och prototypimplementeringar visar vi hur skräddarsydda ML- och systemdesignstrategier kan adressera dessa problem.

Det första målet är att möjliggöra cybersäkerhetsexperiment och detektion av intrång i industriella system. Avhandlingen introducerar ICSSIM, ett nytt ramverk för att skapa flexibla, skalbara och kostnadseffektiva ICS-testbäddar. Vi har dessutom utvecklat ICSFlowGenerator, ett verktyg för att analysera nätverkstrafik och beräkna anpassade nätverksflödesparametrar. Med hjälp av detta verktyg har vi skapat ICS-Flow, en ny datamängd (dataset) för att träna och utvärdera anomalidetekteringsmodeller. ICS-Flow fungerar som en realistisk benchmark för att bedöma ML-baserad intrångsdetektering i ICS-nätverk. 

Med stöd av ICS-Flow adresserar avhandlingen flera tekniska hinder vid driftsättning. För att stödja distribuerade ICS-arkitekturer och minimera beroendet av centrala servrar utforskas en decentraliserad träningsstrategi. Vi undersöker även semi-övervakade inlärningstekniker för att detektera anomalier med enbart normal trafik. För att minska risken för falsklarm integrerar vi ett beslutsstödssystem med IDS:en för att filtrera larm och föreslå åtgärder. Vidare betonar avhandlingen betydelsen av tidsberoende trafikmönster för att identifiera attacktyper och utvärderar effektivitet och resurskrav för olika ML-modeller på ICS-representativ hårdvara. Sammantaget främjar dessa bidrag den praktiska tillämpningen av ML-baserad intrångsdetektering för ICS.

Place, publisher, year, edition, pages
Västerås: Mälardalens universitet, 2025. p. 286
Series
Mälardalen University Press Dissertations, ISSN 1651-4238 ; 440
National Category
Computer Systems Control Engineering Embedded Systems Communication Systems
Research subject
Computer Science
Identifiers
urn:nbn:se:mdh:diva-73132 (URN)978-91-7485-718-4 (ISBN)
Public defence
2025-10-23, Delta och digitalt, Mälardalens universitet, Västerås, 13:30 (English)
Opponent
Supervisors
Available from: 2025-08-28 Created: 2025-08-28 Last updated: 2025-10-10Bibliographically approved

Open Access in DiVA

No full text in DiVA

Authority records

Dehlaghi Ghadim, AlirezaHelali Moghadam, MahshidHansson, Hans

Search in DiVA

By author/editor
Dehlaghi Ghadim, AlirezaHelali Moghadam, MahshidHansson, Hans
By organisation
Embedded Systems
Embedded SystemsComputer SystemsControl Engineering

Search outside of DiVA

GoogleGoogle Scholar

urn-nbn

Altmetric score

urn-nbn
Total: 187 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf