Physics-driven, artificially intelligent virtualization and camouflage for survivable cyber-physical systems
Eventi

Physics-driven, artificially intelligent virtualization and camouflage for survivable cyber-physical systems

02 DICEMBRE 2019

Immagine di presentazione 1

Julian Rrushi
Assistant Professor of Computer Science and Engineering at Oakland University
Michigan, United States

DEIB - Seminar Room "N. Schiavoni" (building 20)
December 2nd, 2019
10.00 am

Contacts:
Stefano Zanero

Research Line:
System architectures

Sommario

We present research on physics-driven operating system and algorithmic mechanisms capable of (i) virtualizing the operating system of industrial control systems with an artificially intelligent hypervisor, which restructures and adapts its own inner workings to match the physics of physical equipment and their physical processes, with the objective of safely collecting high-accuracy security monitoring data at high speed, as well as executing fast agile responses against malware with surgical accuracy; (ii) leveraging the physics-driven architecture of the hypervisor to create physics-compliant cyber decoys along with decoy physical equipment; and (iii) a nonlinear dynamics formalism, rooted in cyber-physical (CPS) decoys and the hypervisor's awareness of physics, which strengthens computational game theory with the objective of modeling and controlling adversarial cyber interactions on CPS environments.



Biografia

Julian Rrushi is an assistant professor of computer science and engineering at Oakland University, Michigan, United States. He is a Ph.D. graduate of the Universita' degli Studi di Milano, class of 2009.
His research is at the intersection of operating systems, computer architecture, and cybersecurity. His research is funded by the U.S.
Department of Defense, and recently also by the U.S. Department of Energy.