A Large-Scale Service-Oriented Sensor Grid Infrastructure

Based on our proposed semantics-based service-oriented model, we are building a large-scale sensor grid infrastructure that can seamlessly integrate heterogeneous sensor resources from different projects distributed across a wide geographical area. This model has several important features. First, it connects heterogeneous sensor resources via the Internet to automatically collect and aggregate sensor data in real-time. Second, the varied types and formats of sensor data are linked to the ontology of the respective sensor resource and the data is converted into a standard ontology-based (OWL) format in order to bind data semantics with the data for ease of access and processing. Third, the sensor data are stored in a large-scale distributed data archive whose schema is also derived from the common concepts found in the ontologies of various sensor nodes. Fourth, the sensor grid integrates computational resources for the compute-intensive processing of sensor data. Fifth, the sensor data can be conveniently accessed and shared via the Web through mash-ups, blogs and Web-services such as Google Earth and Microsoft's SensorMap. We are developing techniques and tools to publish, query, process, visualize, archive and search the vast amount of sensor data efficiently.