Over View
Complex Adaptive Systems (CAS) are a diverse range of natural and engineered systems. However, despite significant progress in their development, there are still major gaps in understanding the resilience of such systems when operating in hostile environments. In this project, we introduce the concept of emergent vulnerabilities, which arise from complex interactions between elements of a CAS. To address this issue, we propose a threat model for CAS and outline several methods for classifying attack surfaces and vectors within these systems.
To quantify resilience to adversarial actions in CAS, we present a Reinforcement Learning-based approach. Our work focuses specifically on smart cities and intelligent transportation systems, as these are examples of emerging CAS. We demonstrate the effectiveness of our proposals in real-world scenarios.
Overall, our project aims to improve understanding of the complex interactions that occur within CAS, and to develop effective strategies for enhancing their resilience to hostile conditions.
Current Team Members:
Dalton Hahn
Vahid Behzadan
Arslan Munir
Affiliate Research Groups:
Intelligent Systems, Computer Architecture, Analytics, and Security lab (Kansas State University)
Tools and Datasets:
Publications:
- Hahn, D. A., Munir, A., & Behzadan, V. (2019). Security and Privacy Issues in Intelligent Transportation Systems: Classification and Challenges. IEEE Intell. Transp. Syst.
- Behzadan, V., & Munir, A. (2018, September). Adversarial Exploitation of Emergent Dynamics in Smart Cities. In 2018 IEEE International Smart Cities Conference (ISC2) (pp. 1-8). IEEE.
- Behzadan, V., & Munir, A. (2017). Models and Framework for Adversarial Attacks on Complex Adaptive Systems. arXiv preprint arXiv:1709.04137.