Unsupervised Attack Isolation in Cyber-physical Systems: A Competitive Test of Clustering Algorithms
KuiZhen Su (),
Chuadhry Mujeeb Ahmed () and
Jianying Zhou ()
Additional contact information
KuiZhen Su: Singapore University of Technology and Design
Chuadhry Mujeeb Ahmed: University of Strathclyde
Jianying Zhou: Singapore University of Technology and Design
Chapter Chapter 3 in Cyberdefense, 2023, pp 27-47 from Springer
Abstract:
Abstract When a complex cyber-physical infrastructure is attacked, operators need to isolate the attack location. Since sensors and actuators are physically intertwined in such structures, operators must be able to separate incoming status data to isolate the precise location of the cyberattack. We let several unsupervised algorithms compete and analyze the extent to which they can provide fast and efficient analysis in order to support operators with this task, using data from the Secure Water Treatment testbed (SWaT), an experimental infrastructure in Singapore that allows us to simulate the behavior of large infrastructure systems. We find that the k-Shape algorithm performs best. This result suggests that unsupervised algorithms can support human operators efficiently even in critical infrastructures with complex sensor data time series.
Date: 2023
References: Add references at CitEc
Citations:
There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.
Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.
Export reference: BibTeX
RIS (EndNote, ProCite, RefMan)
HTML/Text
Persistent link: https://EconPapers.repec.org/RePEc:spr:isochp:978-3-031-30191-9_3
Ordering information: This item can be ordered from
http://www.springer.com/9783031301919
DOI: 10.1007/978-3-031-30191-9_3
Access Statistics for this chapter
More chapters in International Series in Operations Research & Management Science from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().