An ontology design pattern for IoT device tagging systems

Modeling devices has become a crucial task in the Internet of Things (IoT) and Semantic Web technologies are seen as a promising tool for this purpose. However, as it may be arduous to manipulate semantic models, industrial solutions often re-define non-standard, simplified semantics. This is the case with Project Haystack, a framework to tag devices with labels from a predefined vocabulary in the field of Building Automation. In order to make Project Haystack standard and fully semantic, we wrapped its vocabulary in an ontology. In this paper, we present the general structure of this ontology, along with a method to turn tag sets into a Semantic Web model and back. The whole results in a reusable ontology design pattern. We aligned our Haystack tagging ontology with the wide-spreaded Semantic Sensor Network upper ontology and we designed a configuration environment for Building Automation systems based on semantic data to illustrate, so as to discuss the added-value of semantics in automation.