Context-aware service system modeling using timed CPN

In recent years, the advances of computer, network and sensor technologies promote the research and development of context-aware system greatly. Service logics in context-aware system should be correct and reasonable to make sure the system provides user satisfied services. In this paper, we propose a context-aware system modeling methodology based on Timed Colored Petri Net (Timed CPN) to check the correctness of service logics. To simulate the real situation, including the randomness of contexts, we extend CPN by introducing time attributes and random functions. Since many of the service logics are described in ECA rules, we also develop four basic transforming patterns that are used to transform ECA rules into Timed CPN models. The whole system model can be obtained by fusing the Timed CPN models of service logic and context randomness. We describe this methodology in detail by modeling a case of smart home and verify its feasibility and validity.