A comprehensive sensor taxonomy and semantic knowledge representation: Energy meter use case

The increasing use of sensors and their observations in applications like environmental monitoring, security and surveillance, health care, infrastructure, meteorology and others not only generate huge amount of sensor data but also increase complexity of integration of heterogeneous sensor devices, their data formats and procedures of measurements. Therefore ways to manage sensors, sensing devices and systems and thereby handling generation of large volume of sensor data is becoming very important. Formal definition of sensor data encodings and web services to store and access them given by Sensor Web Enablement (SWE) initiative of Open Geospatial Consortium (OGC) provide syntactic interoperability but collecting, reasoning, querying on sensors and their observations require sensor semantic compatibility. It allows users to work with domain concepts, their relations and restrictions, which is an abstraction above the technical nitty-gritty of diverse sensor data format and their integration. The paper describes various sensor concepts and their relationships extending IEEE SUMO upper level ontology and OntoSensor, including SensorML and classifies sensor information into five major sensor knowledge representation (1) hierarchy (2) data (3) function (4) data exchange and (5) domain specific along with code snippets of semantic services generated by mapping between conceptual relationships with structural relationships described in object oriented languages like C++ or Java.