Towards Secure Networked Cyber-Physical Systems: A Theoretic Framework with Bounded Rationality
pdf
Abstract:
Securing critical networked cyber-physical systems (NCPSs) such as the power grid or transportation systems has emerged as a major national and global priority. The networked nature of such systems renders them vulnerable to a range of attacks both in cyber and physical domains as corroborated by recent threats such as the Stuxnet virus. Developing security mechanisms for such NCPSs significantly differs from traditional networked systems due to strong synergies between heterogeneous yet interdependent cyber and physical subsystems and the presence of diverse decision makers. These challenges necessitate rethinking existing security solutions while abstracting the underlying technology and incorporating cyber, physical and human elements. The main goal of this research is to develop the necessary science and engineering tools for designing security solutions tailored to NCPSs. This project develops multidisciplinary framework that weaves together principles from cybersecurity, game theory, networking, and criminology to yield many advances: 1) novel security mechanisms for NCPSs founded on solid game-theoretic and control-theoretic notions, 2) experiments with real-world attackers coupled with new analytical tools that allow to symbiotically incorporate bounded human rationality in NCPS security, and 3) implementation of the theoretical tools over realistic environments such as a newly built, cross-institutional NCPS simulator. This research transcends specific cyber-physical systems domains and provides the necessary tools to building secure and trustworthy NCPSs. The broader impacts include a new infrastructure for NCPS research and education, training of students, new courses, and outreach events focused on under-represented student groups.
Submitted by Walid Saad
on
Abstract:
Securing critical networked cyber-physical systems (NCPSs) such as the power grid or transportation systems has emerged as a major national and global priority. The networked nature of such systems renders them vulnerable to a range of attacks both in cyber and physical domains as corroborated by recent threats such as the Stuxnet virus. Developing security mechanisms for such NCPSs significantly differs from traditional networked systems due to strong synergies between heterogeneous yet interdependent cyber and physical subsystems and the presence of diverse decision makers. These challenges necessitate rethinking existing security solutions while abstracting the underlying technology and incorporating cyber, physical and human elements. The main goal of this research is to develop the necessary science and engineering tools for designing security solutions tailored to NCPSs. This project develops multidisciplinary framework that weaves together principles from cybersecurity, game theory, networking, and criminology to yield many advances: 1) novel security mechanisms for NCPSs founded on solid game-theoretic and control-theoretic notions, 2) experiments with real-world attackers coupled with new analytical tools that allow to symbiotically incorporate bounded human rationality in NCPS security, and 3) implementation of the theoretical tools over realistic environments such as a newly built, cross-institutional NCPS simulator. This research transcends specific cyber-physical systems domains and provides the necessary tools to building secure and trustworthy NCPSs. The broader impacts include a new infrastructure for NCPS research and education, training of students, new courses, and outreach events focused on under-represented student groups.