ORDEN: outlier region detection and exploration in sensor networks

Sensor networks play a central role in applications that monitor variables in geographic areas such as the traffic volume on roads or the temperature in the environment. A key feature users are often interested in when employing such systems is the detection of unusual phenomena, that is, anomalous values measured by the sensors. In this demonstration, we present a system, called ORDEN, that allows for the detection and (visual) exploration of outliers and anomalous events in sensor networks in real-time. In particular, the system constructs outlier regions from anomalous sensor measurements to provide for a comprehensive description of the spatial extent of phenomena of interest. With our system, users can interactively explore displayed outlier regions and investigate the heterogeneity within individual regions using different parameter and threshold settings. Using real-world sensor data streams from different application domains, we demonstrate the effectiveness and utility of our system.