Ontology-based data processing in wireless sensor networks
Stijn Verstichel · Ghent University Academic Bibliography (Ghent University) · 2010
Wireless sensor networks (WSN) consist of a large number of potential heterogeneous sensor nodes, interconnected by means of optimized wireless communication links. Oftentimes, these networks use a hierarchical approach with a dedicated sink node which serves as an interconnection between the resource constrained remote wireless sensor network and the more powerfull back-end applications. To enable intelligent monitoring and data management in such an environment, we have developed a fully distributed ontologybased aggregation framework. This framework has been deployed and evaluated on the WiLab.t wireless sensor and mesh network testbed. However, because of the adoption of a generic ontology approach, the framework could be used on other real life WSNs as well. Apart from offering different kinds of information, the ontology-based approach also allows classification and inference to be performed on the gathered data. As such, only important conclusions are exposed to the back-end application. Additionally, extra precautions have been taken to minimize the network usage between the remote sink node and the back-end services.