A Game Theoretic Approach to Analysis and Design of Survivable and Secure Systems and Protocols
This paper proposes a game theoretic methodology for analysis and design of survivable and secure systems and protocols. Modern game theory provides a natural setup for dealing with adversarial situations, which may involve multiple adversaries with different objectives, resources, access and risk tolerance, e.g., hackers, terrorists, and national intelligence organizations. The solution based on the game theoretic methodology balances ease of access and cost-effectiveness on the one hand, with survivability, fault tolerance, and security on the other hand. The main methodological difficulty in developing game theoretic model of real systems is quantifications of the “rules of the game”, including set of feasible strategies and utility function for each player. Computational challenges result from difficulty of solving games with realistic number of players and feasible strategies.
KeywordsNash Equilibrium Fault Tolerance Admission Control Risk Tolerance Feasible Strategy
Unable to display preview. Download preview PDF.
- 1.Neumann, P.G.: Practical Architectures for Survivable Systems and Networks. Technical report, SRI International (2000), http://www.csl.sri.com/neumann/survivability.pdf
- 2.Salter, C., Saydjari, O.S., Schneier, B., Walner, J.: Towards a Secure System Engineering Methodology. Technical Report.: http://www.counterpane.com/secure-methodology.pdf
- 5.Koutsoupias, E., Papadimitriou, C.H.: Beyond competitive analysis. In: Foundations of Computer Science (1994), (updated version available at http://www.cs.berkeley.edu/christos/
- 6.Marbukh, V.: Network Management under Incomplete Information on the Operational Environment. In: International Symposium on Information Theory and Its Applications (ISITA2000), Honolulu, Hawaii, pp. 637–640 (2000)Google Scholar
- 7.Fudenberg, D., Tirole, J.: Game Theory. MIT Press, Cambridge (2000)Google Scholar
- 8.Marbukh, V.: Optimization of Adaptive Distributed Protocols as a Game. In: ACM SIGMETRICS 2003: First Workshop on Algorithms and Architectures for Self-Managing Systems, San Diego, CA, USA (2003)Google Scholar