Event dashboard: Capturing user-defined semantics events for event detection over real-time sensor data

Sensor networks provide the ability to observe physical phenomena in real-time and provide useful information to help conservation and management of environmental resources. However, sensor data meaning, format and interface heterogeneity are barriers to effective discovery and analysis of events of interest. We propose a web-based user application, the Event Dashboard, which facilitates user capture semantics for events of interest over a sensor network. The Event Dashboard user interface is driven by a set of ontologies, which provide metadata about relevant domain concepts and the sensor network. We utilise the SSN sensor ontology to capture constraints on the sensor metadata. We propose ontology extensions for capturing domain and event semantics using a case study in the water quality domain. Our approach allows the event description to be abstracted from specific interfaces of a sensor network and to be used for querying of sensor data. Event descriptions can subsequently be deployed through a semantic mediator to complex event processing and stream processing implementations over a sensor network.