Dynamic theories of trust for secure agent-based systems

Agent-based systems consist of a collection of agents that interact with each other in dynamic unpredictable, and unreliable environments. A very basic problem regarding security properties of agent-based systems is that whether a message sent by an agent is reliably received by other agents and whether the message received is regarded as reliable in the view of receivers. The problem generally depends on the trust that agents would put in the security mechanisms of the system. In order to support the analysis of various security mechanisms in agent-based systems, it is necessary to systematically investigate formal representation techniques, and provide more generic tools for the specification, and reasoning about trust theories formalizing security mechanisms in agent-based systems. In this paper, we use a temporalised belief logic to show how to establish dynamic trust theories for communication protocols. Such theories provide a foundation for reasoning about properties of agent-based systems operating in dynamic environments. We also outline a decision procedure based on model checking for verifying security properties for trust theories.