Semantic mediation of observation datasets through Sensor Observation Services

This paper describes a first effort for the semantic mediation between heterogeneous environmental observation datasets through the Sensor Observation Service (SOS) standard proposed by the Open Geospatial Consortium. The solution enables application domain experts to provide an ontology with semantic data integration knowledge, which is next combined with data source knowledge during the evaluation of global SOS GetObservation requests. This enables the development of a more general purpose solution that may be adapted to different application domains by just changing the ontology. Besides, users without specific application domain skills and knowledge may now develop new semantically enabled applications. Finally, the design of the framework is based on the well-known Mediator/Wrapper architecture and follows a Local As View data integration approach, which simplifies the incorporation of new datasets without having to change the existing data integration knowledge.