Protecting Critical Infrastructure
Dr. Daniel Barbara
KEY INTERESTS
Machine learning; Data mining; Anomaly detection; Artificial intelligence
APPOINTMENTS/AFFILIATIONS
Professor, Department of Computer Science, George Mason University
ACADEMIC DEGREES
BS, Electrical Engineering, Universidad Metropolitana
MS, Computer Science, Princeton University
PhD, Computer Science, Princeton University
PROTECTING CRITICAL INFRASTRUCTURE
Modern critical infrastructures have begun to transition from fully analog architectures towards hybrid ecosystems that incorporate a myriad of heterogeneous digital devices supporting physical interactions. Despite their value, the galloping integration of digital elements into these older systems has exposed them to a new range of threats (such as the Stucnet malware and the cyber-attack against the supervisory control and data acquisition systems of the Ukranian power grid, etc.). Such incidents attest to the fact that not only are critical infrastructures on attackers' radar, but that they also remain largely unprepared and vulnerable to such attacks. This project works to develop an environment where stakeholders are able to experiment with advanced malware and benchmark new defense strategies that are based on systems that support adversarial learning leveraging recent advances in AI and machine learning algorithms. Such strategies include the development and deployment of enhancements to Named Data Networking to orchestrate application- and network-level defenses that can be pushed to the network edge.