Urban transportation system is a distributed, man-involved and time-variable complex system. So the traffic state analysis is important because it can provide, in a real time way, global view of traffic state in the level of road network, which is very clearly useful in traffic control and route guidance. However, the traffic state analysis involves substantial data and multi-sensor data. So the architecture is important because it can define the mechanism for data management and state analysis. The provided architecture in this paper includes the global data management mechanism, the framework of area traffic state analysis, and the uniform traffic state variable. In global traffic data management, the CORBA-A mechanism combing CORBA, agent and XML for data management is provided, which can deal with the distributed heterogeneous dynamic and substantial traffic data. In the traffic state analysis, the layered framework is provided. The uniform state variable is also defined in the architecture, which is the basis of data fusion for GPS data, loop detector data, map data and other data. The case study shows the provided architecture feasible.
