Smart sensors are increasingly being used to manage and monitor critical urban infrastructures, e.g., for telecommunication, transport, water, or energy networks, as well as for healthcare, crime fighting or smart buildings. Sensor-based monitoring systems offer ways of continuously monitoring low frequency activities, and open the door to new analytic and predictive applications in Smarter Cities.
In collaboration with the new IBM Research Smarter Cities Center in Dublin, the eXascale Infolab carries research spanning from anomaly detection in water distribution networks to low-latency analytics for timeseries. We take a bigdata approach in solving those problems by combining state of the art platforms (e.g. Storm, Hadoop, SciDB) to design architectures tailored to each case.
The figure bellow depicts a water distribution network data management architecture. The architecture has three main components: simple water sensors that periodically broadcast their measurements; self-organizing base stations that gather the sensor readings and clean them using a stream-processing flow (on the right) and share them through an overlay network; and an array data management back end that durably stores and analyzes all values.