The challenges of constructing a sensor web data stream management system

The world-wide sensor web provides a powerful monitoring capability for users to monitoring the physical world at spatial and temporal scales that were impossible in the past. With the advance of sensor technologies and the definition of interoperable open standards, the sensor web connects various types of sensors located worldwide and performs observations at high frequency. By linking the large amount of sensor data streams produced by the sensor web, scientists can observe phenomena that were previously unobservable. In order to harness the full potential of the continuous monitoring power of the sensor web, efficiently processing sensor web data streams and providing timely notifications for time-critical events (e.g., disasters or security breaches) are necessary. While there have been general-purpose systems proposed to handle continuous data streams (e.g., data stream management systems), we observed that there are unique challenges caused by the nature of geospatial sensor web data and sensor web data sources. Therefore, in this paper, we present our observations on the challenges of constructing a world-wide sensor web data stream management system.