Interoperability Method Based On Ontologies And LOD In Semantic Web Of Things

Internet of Things (IoT) connects sensors, controllers, humans and a series of things according to the standard communication protocols. It achieves remote control and intelligentialize through interoperability. However, the heterogeneity among things hinders the IoT from achieving interoperability at a global scale. Semantic Web of Things (SWoT) is proposed to solve the problems in IoT that there are a variety of expressions in the client side when the same object uploads to the Internet and the machine does not understand the semantics of the objects. It adopts Semantic Web technology into the IoT. Semantic interoperability means that the things with semantic annotations can access and clearly resolve mutual data to achieve semantics sharing, information integration or knowledge inference.In terms of current problems in SWoT, mainly include un-completeness and non-sharedness of its semantic system which makes it hard to support semantic interoperability between different domains, this paper proposed a method of semantic interoperability called SIMOL (Semantic Interoperability Method based on Ontologies and LOD), which combined the advantages of multi-domain ontologies and Linked Open Data (LOD) in semantic expressing. Firstly, the method annotated the newly-deployed sensors in semi-automatic way and added related semantics to corresponding sensor data, so that machines could understand the semantics about data and reason on them. Secondly, it managed same-domain ontologies in a manner of ontology-entity relational table to get formulism and consensus on concepts in ontologies. Thirdly, it obtained related instances of named entities based on LOD, so their information could be enriched.A semantic interoperability system was designed and developed according to the proposed method. First the overall structure of the system was designed and the first level data flow diagram and each modular data flow diagram were given. Then the running sequence diagram of the system was designed. At last, the SIMOL was compared with current existing method. The result of the experiment shows that SIMOL proposed in this paper can reduce the heterogeneity of sensors on a certain extent by understanding the attributes of sensors and the meaning of the measured data, and provide a more standardized and comprehensive semantics to improve the precision ratio and recall ratio when compared with the traditional method based on single ontology or LOD. Then the interoperability of SWoT can be better implemented based on semantics.