Multi-level anomaly detection in industrial control systems via package signatures and LSTM networks
- Submitting institution
-
Cardiff University / Prifysgol Caerdydd
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 101633957
- Type
- E - Conference contribution
- DOI
-
10.1109/DSN.2017.34
- Title of conference / published proceedings
- IEEE/IFIP International Conference on Dependable Systems and Networks (DSN)
- First page
- 261
- Volume
- 0
- Issue
- -
- ISSN
- 2158-3927
- Open access status
- Technical exception
- Month of publication
- August
- Year of publication
- 2017
- URL
-
http://dx.doi.org/10.1109/DSN.2017.34
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
2
- Research group(s)
-
C - Cybersecurity, privacy and human centred computing
- Citation count
- 33
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This paper was presented at the conference DSN'17 with an acceptance rate of 22.3%. It proposed the first anomaly detector for industrial control systems using deep learning techniques and provided state-of-the-art performance. This work is recognised as the first use of LSTM for ICS-specific anomaly detection (e.g., Virginia Tech: https://arxiv.org/abs/1805.00074 and Tsinghua University: https://arxiv.org/pdf/1802.03903.pdf). This research was supported by the project "RITICS: Trustworthy Industrial Control Systems", co-funded by EPSRC and NCSC (EP/L021013/1: £756k) and EPSRC project “Security by Design for Interconnected Critical Infrastructures” (EP/N020138/1: £203k)
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -