A Generic Recursive Multiagent Model to Simplify Large Scale Multi-level Systems Observation

Observation of large scale complex systems is difficult because of their extension and openness. A multi-scale organization for these systems can be a solution to make them more accessible. This paper proposes a multi-level multi-agent mechanism based on recursion for management, observation and integration of large scale complex systems. It is developed as a decentralized application free framework which allows to real MAS to communicate with agents on abstract layers. As an illustration, the framework is applied to a wireless sensor network supervision system.