Many sources of sensor data exist on campus and the surrounding area. There are several weather stations and solar arrays on campus (danaroofsolar.bucknell.edu). The environmental center has several water quality sensors. The USGS also has different water quality sensors. Campus buildings have at least one energy meter and many zone temperature sensors. All of this data is available online through various web services, however, they are all different. There is no automated way to compare data from different sensing systems . We need a web application that allows many different data sources to be configured and transformed to a common data format.
- A web application which supports users with assignable access levels.
- Permitted users can add a new sensor data source and provide code to query and transform the foreign data source to a common format (to be defined).
- Another configurable service will transform data from the common format to various web visualization engines for plotting.
- Plots can be created via the web interface which query any of the configured sensors using the common data format and appropriate plot converter. Raw data for each plot can also be easily downloaded in csv/xls format for permitted users.
- The configuration (series/y axis/ labels/etc) of each plot can be saved to the user’s profile. Several plots can be dynamically configured to create dashboards. Navigation between the dashboards and plots can also be user defined. Dashboards and plots can then be shared with other users with view/edit rights and also to the public for view.
- Data will not be persisted in this system long term (memory caching is OK).
- On each request (if not cached) the server will contact the remote data source and transform the data.
- Reliable and accurate people counting on a reasonable budget.
- Should work in all weather conditions (day (bright)/ night (dark), rain, snow, etc).
- Robustly handle unusual cases such as large backpacks, carrying boxes, pushing a cart, large groups (e.g., campus tour, fire alarm).
- MIT or GPL
Point of Contact
- Prof. Alan Marchiori, Dept. of Computer Science, Bucknell University