Sensor and Actuator Attacks on Hierarchical Control Systems with Domain-Aware Operator Theory

Conference · Wed Oct 18 04:00:00 EDT 2023

DOI:https://doi.org/10.1109/RWS58133.2023.10284668· OSTI ID:2331405

Bakker, Craig KR ^[1]; Vasisht, Soumya S. ^[1]; Huang, Sen ^[1]; Vrabie, Draguna L. ^[1]

---

1. BATTELLE (PACIFIC NW LAB)

Cyber-Physical Systems (CPSs) provide opportunities for cyber attacks to have physical impacts. Advanced Persistent Threats (APTs) are a subclass of cyber threats that act stealthily to avoid detection and enable long-term attacks. Here, we build on our past work in APT modelling to combine deception-based sensor bias attacks and direct actuator manipulations in attacks against a hierarchical control system. That past work used the Koopman operator to develop a data-driven, domain-aware, optimization-based attacker model. Using an expansion of this model, we compute several different attacks, including multiple simultaneous attacks, against a high-fidelity commercial building emulator and compare the impacts of those attacks to each other. One next step of interest is to construct a defender system, built on the same modelling approach, designed to detect and mitigate such attacks.

• Research Organization: Pacific Northwest National Laboratory (PNNL), Richland, WA (United States)
• Sponsoring Organization: USDOE
• DOE Contract Number: AC05-76RL01830
• OSTI ID: 2331405
• Report Number(s): PNNL-SA-184521
• Country of Publication: United States
• Language: English

Related Subjects

koopman operator
Cyber-Physical Security
machine leaning