Abstract | ||
---|---|---|
Adversary emulation is an offensive exercise that provides a comprehensive assessment of a system’s resilience against cyber attacks. However, adversary emulation is typically a manual process, making it costly and hard to deploy in cyber-physical systems (CPS) with complex dynamics, vulnerabilities, and operational uncertainties. In this paper, we develop an automated, domain-aware approach to adversary emulation for CPS. We formulate a Markov Decision Process (MDP) model to determine an optimal attack sequence over a hybrid attack graph with cyber (discrete) and physical (continuous) components and related physical dynamics. We apply model-based and model-free reinforcement learning (RL) methods to solve the discrete-continuous MDP in a tractable fashion. As a baseline, we also develop a greedy attack algorithm and compare it with the RL procedures. We summarize our findings through a numerical study on sensor deception attacks in buildings to compare the performance and solution quality of the proposed algorithms. |
Year | DOI | Venue |
---|---|---|
2020 | 10.1109/ISI49825.2020.9280521 | 2020 IEEE International Conference on Intelligence and Security Informatics (ISI) |
Keywords | DocType | ISBN |
Adversary Emulation,Reinforcement Learning,Cyber-Physical Security,Hybrid Attack Graph | Conference | 978-1-7281-8801-0 |
Citations | PageRank | References |
0 | 0.34 | 6 |
Authors | ||
5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Arnab Bhattacharya | 1 | 0 | 0.34 |
Thiagarajan Ramachandran | 2 | 1 | 1.74 |
Sandeep Banik | 3 | 0 | 0.68 |
Chase P. Dowling | 4 | 0 | 0.34 |
Shaunak D. Bopardikar | 5 | 1 | 5.48 |