Explainable and perturbation-resilient model for cyber-threat detection in industrial control systems Networks
Abstract Deep learning-based intrusion detection systems (DL-IDS) have proven effective in detecting cyber threats. However, their vulnerability to adversarial attacks and environmental noise, particularly in industrial settings, limits practical application. Current IDS models often assume ideal co...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2025-02-01
|
| Series: | Discover Internet of Things |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s43926-025-00100-0 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832571350117515264 |
|---|---|
| author | Urslla Uchechi Izuazu Cosmas Ifeanyi Nwakanma Dong-Seong Kim Jae Min Lee |
| author_facet | Urslla Uchechi Izuazu Cosmas Ifeanyi Nwakanma Dong-Seong Kim Jae Min Lee |
| author_sort | Urslla Uchechi Izuazu |
| collection | DOAJ |
| description | Abstract Deep learning-based intrusion detection systems (DL-IDS) have proven effective in detecting cyber threats. However, their vulnerability to adversarial attacks and environmental noise, particularly in industrial settings, limits practical application. Current IDS models often assume ideal conditions, overlooking noise and adversarial manipulations, leading to degraded performance when deployed in real-world environments. Additionally, the black-box nature of DL model complicates decision-making, especially in industrial control systems (ICS) network, where understanding model behavior is crucial. This paper introduces the eXplainable Cyber-Threat Detection Framework (XC-TDF), a novel solution designed to overcome these challenges. XC-TDF enhances robustness against noise and adversarial attacks using regularization and adversarial training respectively, and also improves transparency through an eXplainable Artificial Intelligence (XAI) module. Simulation results demonstrate its effectiveness, showing resilience to perturbation by achieving commendable accuracy of 100% and 99.4% on the Wustl-IIoT2021 and Edge-IIoT datasets, respectively. |
| format | Article |
| id | doaj-art-001ba2f44e6f41ba91e837d520e2e105 |
| institution | Kabale University |
| issn | 2730-7239 |
| language | English |
| publishDate | 2025-02-01 |
| publisher | Springer |
| record_format | Article |
| series | Discover Internet of Things |
| spelling | doaj-art-001ba2f44e6f41ba91e837d520e2e1052025-02-02T12:37:37ZengSpringerDiscover Internet of Things2730-72392025-02-015112310.1007/s43926-025-00100-0Explainable and perturbation-resilient model for cyber-threat detection in industrial control systems NetworksUrslla Uchechi Izuazu0Cosmas Ifeanyi Nwakanma1Dong-Seong Kim2Jae Min Lee3Institut fur Datentechnik und Kommunikationsnetze, Tchnische Universitat Carolo-Wilhelmina zu, Hans-Sommer-Strasse 66Lane Department of Computer Science and Electrical Engineering, West Virginia UniversityInstitut fur Datentechnik und Kommunikationsnetze, Tchnische Universitat Carolo-Wilhelmina zu, Hans-Sommer-Strasse 66Institut fur Datentechnik und Kommunikationsnetze, Tchnische Universitat Carolo-Wilhelmina zu, Hans-Sommer-Strasse 66Abstract Deep learning-based intrusion detection systems (DL-IDS) have proven effective in detecting cyber threats. However, their vulnerability to adversarial attacks and environmental noise, particularly in industrial settings, limits practical application. Current IDS models often assume ideal conditions, overlooking noise and adversarial manipulations, leading to degraded performance when deployed in real-world environments. Additionally, the black-box nature of DL model complicates decision-making, especially in industrial control systems (ICS) network, where understanding model behavior is crucial. This paper introduces the eXplainable Cyber-Threat Detection Framework (XC-TDF), a novel solution designed to overcome these challenges. XC-TDF enhances robustness against noise and adversarial attacks using regularization and adversarial training respectively, and also improves transparency through an eXplainable Artificial Intelligence (XAI) module. Simulation results demonstrate its effectiveness, showing resilience to perturbation by achieving commendable accuracy of 100% and 99.4% on the Wustl-IIoT2021 and Edge-IIoT datasets, respectively.https://doi.org/10.1007/s43926-025-00100-0Cyber-securityDeep learningAdversarial attackIndustrial control systemXAIIntrusion detection |
| spellingShingle | Urslla Uchechi Izuazu Cosmas Ifeanyi Nwakanma Dong-Seong Kim Jae Min Lee Explainable and perturbation-resilient model for cyber-threat detection in industrial control systems Networks Discover Internet of Things Cyber-security Deep learning Adversarial attack Industrial control system XAI Intrusion detection |
| title | Explainable and perturbation-resilient model for cyber-threat detection in industrial control systems Networks |
| title_full | Explainable and perturbation-resilient model for cyber-threat detection in industrial control systems Networks |
| title_fullStr | Explainable and perturbation-resilient model for cyber-threat detection in industrial control systems Networks |
| title_full_unstemmed | Explainable and perturbation-resilient model for cyber-threat detection in industrial control systems Networks |
| title_short | Explainable and perturbation-resilient model for cyber-threat detection in industrial control systems Networks |
| title_sort | explainable and perturbation resilient model for cyber threat detection in industrial control systems networks |
| topic | Cyber-security Deep learning Adversarial attack Industrial control system XAI Intrusion detection |
| url | https://doi.org/10.1007/s43926-025-00100-0 |
| work_keys_str_mv | AT ursllauchechiizuazu explainableandperturbationresilientmodelforcyberthreatdetectioninindustrialcontrolsystemsnetworks AT cosmasifeanyinwakanma explainableandperturbationresilientmodelforcyberthreatdetectioninindustrialcontrolsystemsnetworks AT dongseongkim explainableandperturbationresilientmodelforcyberthreatdetectioninindustrialcontrolsystemsnetworks AT jaeminlee explainableandperturbationresilientmodelforcyberthreatdetectioninindustrialcontrolsystemsnetworks |