CANCELLED - SPAN: A new paradigm for collecting and processing field data using real-time in-situ sensing, data models, and workflows, enabling long term data-driven science

Background/Question/Methods   We present a new system that provides a flexible, vendor neutral, embedded cyber infrastructure for environmental sensing. This technology, which has been fielded in a dozen deployments worldwide, enables rapid integration of multiple sensors, data acquisition units, autonomic programmable sensor to sensor triggering, QA/QC, multiple data communication protocols (including satellite, WiFi, long-range radios, cable modem etc.) and several web based Service Oriented Architecture (SOA) data analysis interfaces including a Google Cloud application. This technology transforms many aspects of field science through its programmable low power embedded processor in the field which not only provides the capabilities described above, but through additional R&D has the potential to provide a rich environment for automatic metadata tagging, data repository submission, and data sharing. This work is part of a continuing series of embedded cyber infrastructure efforts under the umbrella of the SPAN project at The University of Southern California's Information Sciences Institute.

Results/Conclusions   SPAN is designed to be turnkey and easy to use by scientists. It provides a sensor integration framework that includes support for a variety of traditional and emerging sensors such as microphones, webcams, and robotic cameras. Many of the benefits of the system are achieved through the use of a powerful, low-cost, low-power generic embedded processor close to the sensors in the field. The use of multiple communication technologies provide real-time data access to scientists in the laboratory and enable two-way communication to the field, which can be used for remote sensor reconfiguration capability without requiring field visits! SPAN's integrated powerful processor also allows the quality control and assurance of collected data as well as the incorporation of rich and complex data models and workflow technologies in the field. For instance, scientists can receive e-mails or text messages if there are problems in the remote site, or if an event of interest is detected. The integration of SPAN with compute and data workflow engines is an ongoing effort. The SPAN system provides a bridge from the field scientist to long term data analysis and evaluation frameworks.