Semantic Interoperability in Body Area Sensor Networks and Applications

Crucial to the success of Body Area Sensor Networks is the flexibility with which stakeholders can share, extend and adapt the system with respect to sensors, data and functionality. The first step is to develop an interoperable platform with explicit interfaces, which takes care of common management tasks. Beyond that, interoperability is defined by semantics. This paper presents the analysis, design, implementation and evaluation of a semantic layer within an existing BASN platform for the purpose of improving the semantic interoperability among sensor networks and applications. We adopt an ontology-based approach but rather than having a single overall ontology, we find that using clear semantic domains and mappings between them improves composability and reduces interoperability problems. We discuss the design choices and a reference implementation on an Android phone and actual sensor devices. We show by a qualitative evaluation that this semantic interoperability indeed provides significant improvements in flexibility.