BLOC: A Game Theoretic Approach to Orchestrate CPS Against Cyber Attacks
| dc.contributor.advisor | Guirguis, Mina | |
| dc.contributor.author | Tahsini, Alireza | |
| dc.contributor.committeeMember | Tešić, Jelina | |
| dc.contributor.committeeMember | Gu, Qijun | |
| dc.date.accessioned | 2021-09-14T14:17:32Z | |
| dc.date.available | 2021-09-14T14:17:32Z | |
| dc.date.issued | 2019-08 | |
| dc.description.abstract | One important aspect in protecting CPS is ensuring that the proper control and measurement signals are propagated within the control loop. The CPS research community has been developing a large set of check blocks that can be integrated within the control loop to check signals against various types of attacks (e.g., false data injection attacks). Unfortunately, it is not possible to integrate all these "checks" within the control loop as the overhead in checking signals may violate the delay constraints dedicated by the control loop. Moreover, these blocks do not completely operate in isolation of each other, but dependencies exist among them in terms of their effectiveness against detecting a subset of attacks. Thus, it becomes a challenging and a complex problem to assign the proper checks, specially with the presence of a rational adversary who can observe the check blocks assigned and optimizes her own attack strategies accordingly. This paper tackles the inherent state-action space explosion that arise in securing CPS through developing a unifying framework in which Deep Reinforcement Learning algorithms are utilized to provide optimal/sub-optimal assignments of check blocks to signals. The framework models stochastic games between the adversary and the CPS defender and derives mixed strategies for assigning check blocks to ensure the integrity of the propagated signals, while abiding to the real-time constraints dictated by the control loop. Furthermore, the strategies obtained reflect various factors such as the aggressiveness and risk associated to the players in taking defense/attack actions. Our results show that our framework can obtain assignment strategies that outperform other strategies and heuristics. | |
| dc.description.department | Computer Science | |
| dc.format | Text | |
| dc.format.extent | 43 pages | |
| dc.format.medium | 1 file (.pdf) | |
| dc.identifier.citation | Tahsini, A. (2019). <i>BLOC: A game theoretic approach to orchestrate CPS against cyber attacks</i> (Unpublished thesis). Texas State University, San Marcos, Texas. | |
| dc.identifier.uri | https://hdl.handle.net/10877/14494 | |
| dc.language.iso | en | |
| dc.subject | Deep reinforcement learning | |
| dc.subject | Deep learning | |
| dc.subject | Game theory | |
| dc.subject | Markov game | |
| dc.subject | CPS security | |
| dc.subject | Cyber physical systems | |
| dc.title | BLOC: A Game Theoretic Approach to Orchestrate CPS Against Cyber Attacks | |
| dc.type | Thesis | |
| thesis.degree.department | Computer Science | |
| thesis.degree.discipline | Computer Science | |
| thesis.degree.grantor | Texas State University | |
| thesis.degree.level | Masters | |
| thesis.degree.name | Master of Science |