Ultra-scale environmental monitoring - the intelligent river

Sensor network deployments monitor a wide range of properties in real-time, providing data streams describing a host of environments. The volume of data in these real-time streams can quickly overwhelm monitoring systems, communication infrastructure, and downstream processing system. To address these issues, we designed the Intelligent River, a three layer architecture consisting of a sensor fabric, a data communication network, and a data analytics system. Our sensor fabric can be adapted for use in multiple applications and environments by changing communication and sensor configurations. Wireless mesh networks for dense deployments, and cellular networks for remote deployments handle the back-end communication. The analytics system involves multiple processing queues for collecting and processing raw data, and a web-based portal presentation. The web portal provides modules to register sensor information with the cloud infrastructure, view real-time and historical data, and monitor the status of the deployments. In this demo, we will be presenting the end-to-end system through a physical testbed. We will also showcase our real time data-streams originating from various sensor fabrics deployed across various environments.