We were inspired by the data-sharing philosophy of Microshare to create a freely-available, public-good data set that could leverage raw data streams from commercial, government and community sources. We chose air quality, especially those aspects that directly impact residents with respiratory ailments, as our focus because it is sensor-intensive information to collect, and the correlation to geographical location adds a significant layer of value to the individual raw sources.
What it does
Open Air collects air quality data from numerous open data streams, focusing on particulates and other respiratory-aggravating gasses, and correlates that data onto a Goolge Maps overlay. The resulting composite data set could be used: directly by consumers as a go / no-go map, to generate area-wide pollution alerts, to identify pollution sources for mitigation, as a long-term analytics data source etc.
How we built it
With Microshare as a data-sharing hub, Open Air can leverage data streams from numerous sources. Commercial business that monitor air quality for their own enterprise needs could share certain portions of their data with Open Air. Governmental agencies could contribute their own sensor data as well supplement their data with outside streams. Citizen-scientists could contribute data from their own, personal sensor installations and, finally, consumers could access the data directly to help plan their daily activities and to be aware of potentially dangerous conditions.
Challenges we ran into
We ran into a number of challenges in getting data from sensors, through our development boards, over the HackIoT network provider - Senet - and into the Microshare platform. This was largely due to our own lack of experience and we believe we could do it more quickly in the future. Our team was also lacking in web app developers, so we had to rely on off-the-shelf dashboards, but we know that a skilled web app development team could do so much more.
Accomplishments that we're proud of
Successfully completing the HackIoT tutorials: streaming data from our dev boards, through Senet to Microshare, then successfully sharing our individual data streams with our teams.
What we learned
The generic tutorials clearly illustrated concepts that could be applied to the specific sensors we would ultimately need to effectively demonstrate our concept. We also learned that our team would need additional web app development talent to fully demonstrate our vision.
Underlying hardware technology:
Sensor readings of interest: PM2.5 and PM10 Particulates Sulfur Dioxide Carbon Monoxide Nitrogen Dioxide Ozone (each one needs a different sensor)
Air Pollution Index (API) https://en.wikipedia.org/wiki/Air_Pollution_Index
Info on air quality measurements http://www.livefrombeijing.com/2008/08/what-is-the-api-and-how-is-it-calculated/
Example of Air pollution Database for beijing, Microshare will be our database/datastore https://qz.com/197786/six-years-of-bejing-air-pollution-summed-up-in-one-scary-chart/