Developing Research Capability in Cyber-Physical Systems at the University of Florida
Abstract
Cyber-physical systems (CPS) represent some of the most pervasive and important computation frameworks found in modern society. From controlling new and innovative technologies such as the smart grid, to providing health and life support through implantable medical devices, to supporting the nationÕs critical infrastructure, understanding how these systems work and securing them from our enemies is of paramount importance to the nation and its residents. Cyber-physical systems tightly integrate computation and physical processes. They are typically designed as a network of computational elements interacting with physical input and output instead of as standalone components. The computation engine reads inputs from sensors, processes the data then emits actions through actuators in a continuous latency-critical loop. Unlike traditional embedded systems, where the emphasis is on the computational units of highly application-specific closed systems, the focus in CPS is on the interaction between computation and physical processes. Whole system analysis and control feedback are thus critical differentiators that much be considered when analyzing CPS. Additionally, these systems are inter-dependent in such a way that changing in one network may have a profound impact on other systems. One notable example of such interdependent networked system is the smart-grid, of which the power stations depend on communication networks for control and management and vice versa, the communication networks depend on the power systems for their electricity supports. As these systems connect with cyberspace, and the Internet in particular, these interface points open new surfaces of vulnerabilities and threats vectors. The challenge is to understand the operations of these systems, their interdependencies, and the means by which attackers can exploit them, in order to provide robust models and systems to defend this critical infrastructure. This DURIP project provides equipment that would extend the capabilities of the University of Florida to support rigorous and realistic testing modeling and analysis of CPS operation, as well as attacks and defenses in these environments. In particular, the equipment would allow this research team to perform real experiments without interfering or damaging commercial networks. Through a combination of real components used in the field with high fidelity simulation environments, we can provide realistic scenarios and to model this data, store the simulation and emulation data for replay and analysis, and incorporate a level of realism that has been impossible to replicate in published academic research to date. A CPS testbed further advances the University of FloridaÕs academic mission, as it is a means for acquiring advanced skills and knowledge about real-world systems and their security for undergraduate and graduate students. Such a testbed can form the basis for modules in courses relating to embedded systems, hardware, network and systems security at varying degrees of sophistication, from introduction to security courses to advanced courses in embedded systems security. In particular, many engineering curricula have courses that focus on case studies of past disasters based on failures of design and process. We believe that these lessons can be vividly brought to life through demonstrating through a testbed how they can occur within CPS. The project has supported efforts in building models for the grid, telecommunications, and embedded research.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Apr 01, 2019
- Source ID
- W911NF1510507
Entities
People
- Kevin Butler
Organizations
- Army Contracting Command
- United States Army
- University of Florida