Semantic surface representation of physical entity in the WEB of things

The Web of Things is a refinement of the Internet of Things, which reuses existing web technologies, standards, languages and tools to integrate the physical world and virtual world seamlessly. Architectures enabling a Web of Things have been proposed, but machine-readable and machine-interpretable approach to provision trillions of heterogeneous available physical entities in virtual world has not acquired its due emphasis. We focus on semantic surface representations of physical entities which are created using HTML, described using a metadata model and formatted using microformat, RDFa and microdata. The interlinked metadata model we proposed consists of three components: physical entity, resource and service. We analyze their relationships and present important attributes in the form of 2-tuple 'metadata, value type' to construct the metadata model. In order to make heterogeneous physical entities searchable and accessible on a large scale and guarantee surface representations machine-readable and machine-interpretable, we introduce microformat, RDFa and microdata to format and add semantics to surface representations. Moreover, a comparison of them is presented to help developers to choose between them.