Sense and Sens'ability: Semantic Data Modelling for Sensor Networks

Sensor networks are used in various applications in several domains for measuring and determining physical phenomena and natural events. Sensors enable machines to capture and observe characteristics of physical objects and features of natural incidents. Sensor networks generate immense amount of data which requires advanced analytical processing and interpretation by machines. Most of the current efforts on sensor networks are focused on network technologies and service develop- ment for various applications, but less on processing the emerging data. Sensor data in a real world application will be an integration of various data obtained from different sensors such as temperature, pressure, and humidity. Processing and interpretation of huge amounts of heterogeneous sensor data and interoperability are important issues in designing a scalable sensor network architecture. This paper describes a semantic model for heterogeneous sensor data representation. We use common standards and logical description frameworks proposed by the Semantic Web community to create a sensor data description model. The work describes a sensor data ontology which is created based on the Sensor Web Enablement (SWE) and SensorML data component models. We describe how the semantic relationship and operational constraints are deployed in a uniform structure to describe the heterogeneous sensor data.