Design and analysis of Bayesian game in role based trust management

Role based trust management is capable of handling the dynamicity and uncertainty of security in multi-domain distributed environment. The interactions between trust domains and users can be modeled as Bayesian games in which both players have their own strategies and payoffs. This paper shows how Game Theory can be a tool to analyze the behaviors of every player in role based trust framework. We describe the types of players and design the details of games. Considering two types of users - cooperative and malicious, we analyze the strategy sets and payoffs of trust domains and each type of users. The equilibriums of the game show how a trust domain can select proper strategies to maximize its total profits.