Covert Cognizance: Embedded Intelligence for Industrial Systems
Can a critical industrial system, such as a nuclear reactor, be made self-aware and cognizant of its operational history? Can it alert authorities covertly to malicious intrusion without exposing its defense mechanisms? What if the intruders are highly knowledgeable adversaries, or even insiders that may have designed the system? This thesis addresses these research questions through a novel physical process defense called Covert Cognizance (C2).
C2 serves as a last line of defense to industrial systems when existing information and operational technology defenses have been breached by advanced persistent threat (APT) actors or insiders. It is an active form of defense that may be embedded in an existing system to induce intelligence, i.e., self-awareness, and make various subsystems aware of each other. It interacts with the system at the process level and provides an additional layer of security to the process data therein without the need of a human in the loop.
The C2 paradigm is founded on two core requirements – zero-impact and zero-observability. Departing from contemporary active defenses, zero-impact requires a successful implementationto leave no footprint on the system ensuring identical operation while zero-observability requires that the embedding is immune to pattern-discovery algorithms. In other words, a third-party such as a malicious intruder must be unable to detect the presence of the C2 defense based on observation of the process data, even when augmented by machine learning tools that are adept at pattern discovery.
In the present work, nuclear reactor simulations are embedded with the C2 defense to induce awareness across subsystems and defend them against highly knowledgeable adversaries that have bypassed existing safeguards such as model-based defenses. Specifically, the subsystems are made aware of each other by embedding critical information from the process variables of one sub-module along the noise of the process variables of another, thus rendering the implementation covert and immune to pattern discovery. The implementation is validated using generative adversarial nets, representing a state-of-the-art machine learning tool, and statistical analysis of the reactor states, control inputs, outputs etc. The work is also extended to data masking applications via the deceptive infusion of data (DIOD) paradigm. Future work focuses on the development of automated C2 modules for “plug ‘n’ play” deployment onto critical infrastructure and/or their digital twins.
History
Degree Type
- Doctor of Philosophy
Department
- Nuclear Engineering
Campus location
- West Lafayette