SEPSen: Semantic event processing at the sensor nodes for energy efficient wireless sensor networks

Abstract

Traditionally in WSNs, the sensor nodes are used only for capturing data that is then later analyzed in the more powerful gateway nodes. This requires a continuous communication that wastes energy at the sensor nodes and greatly reduces the overall network lifetime. We propose a semantic-based in-network data processing that reduces energy consumption and improves the scalability of heterogeneous sensor networks. Ontology fragments in each sensor node help identify the data routed through the sensor network. We have adapted a matching algorithm to process a changing knowledge base. Simulation results show that the networks' energy consumption is considerably reduced.