Pacific Northwest National Lab
Hybrid Attack Graph for Cyber-Physical System Resilience Experimentation
Dr. Sumit Purohit is a Data Scientist at Pacific Northwest National Laboratory (PNNL), Physical and Computational Science Directorate. His research work focuses on scalable graph algorithms, temporal graph analytics, knowledge graphs, and geometric deep learning while working on various DARPA and DOE projects in cybersecurity, national security, and social network domains. He is a co-investigator on the PNNL DARPA Modeling Adversarial Activity (MAA) project and PI on an internal Laboratory-Directed Research and Development (LDRD) project that explores hybrid attack graphs (HAGs) for cyber physical system such as microgrids to support resiliency assessment experimentations.
Hybrid attack graphs provide a flexible and efficient approach to generate attack sequences for a Cyber Physical System (CPS), where CPS state-dynamics is represented as nodes and adversary tactics, and physical actions are represented as edges that cause the system to change from one state to another. The talk will present novel theory and algorithms to generate hybrid attack graphs (HAGs) for cyber-physical system (CPS) resilience experimentation at desired scale and speed.