An Architecture for Enabling IoT Edge Devices to Allow Scalable Publishing of Semantic Linked Data

Abstract

The Internet of Things (IoT) is growing exponentially and is creating enormous sets of data that often reside in proprietary formats and storage containers. Linked Data and the Semantic Web has been around for a long time but adoption by industry has been slow. Linked Data research tends to have an academic bias and real world applicability is often neglected. Based on our experience with ubiworx, an IoT framework used in several real world projects, we propose an architecture using intelligent edge gateways, scalable cloud micro-services and semantic adaption layers to bridge the gap between legacy sensor data acquisition and publishing linked data to ontologies such as Semantic Sensor Network (SSN) and Smart Appliances REFerence ontology (SAREF). In combination, the edge gateway paradigm and the layered cloud micro-services paradigm intersect to provide an accelerated path to publishing linked sensor data while achieving the scalability required by the explosive growth of the IoT and doing so with the best efficiency of resources. Dynamically creating Linked Data serialisations from rapidly changing sensor data sets is a key enabler for adoption of linked data principals in the IoT space.