Science of Secure and Resilient Cyber-Physical Systems
Abstract
The objective of the project was to develop the Science of Secure and Resilient Cyber-Physical Systems. In the area of hierarchical coordination and control, the project developed tools for threat and attack modeling in order to understand the impact of cyber-attacks to system performance and operation, methods to improve resilience in CPS monitoring, resilient distributed control and coordination algorithms in the presence of malicious agents, and algorithms for improving the resilience and security of CPS flow networks such transportation, water, and power networks. In the area of decentralized security, the project developed methods for secure decentralized control, investigated in-depth multi defender games and high-resolution multi-stage security games. Our work investigated practical reasoning techniques for addressing security problems that affect CPS including actor-networks as a formal model of computation, malware detection using active learning, adversarial machine learning, privacy in CPS, optimal personalized filtering, and methods for human-in-the-loop visual analytics for exploring document collections. The project developed a suite of simulation and hardware-in-the-loop testbeds and conducted comprehensive evaluation of both system and security properties. Finally, outreach and education activities included summer camps for high school students and community outreach supporting multiple conferences and workshops and the science of security virtual organization (SOS-VO).
Document Details
- Document Type
- Technical Report
- Publication Date
- Feb 25, 2020
- Accession Number
- AD1092617
Entities
People
- Xenofon Koutsoukos
Organizations
- Vanderbilt University