TOWARDS A LOGICAL BASIS FOR COMMUNICATION IN NETWORK MANAGEMENT

Well-defined interaction between agents is essential if they are to operate successfully in complex, dynamic domains such as network management. In order to co-operate to achieve the goals of the domain, agents must be able to reason about the beliefs and intentions of other agents. In this paper, the authors argue that logic can provide the means by which multiagent interaction can be defined. An extension to first order logic is suggested which will provide the mechanisms necessary to allow agents to reason about the beliefs and intentions of other agents. Use of this logic is illustrated using an example problem from the network traffic management domain.