16 comments

[ 4.5 ms ] story [ 58.0 ms ] thread
This is very similar to pressureNET, my open source Android app that collects pressure data and sends it to atmospheric science researchers. I'm quite excited that there's competition and that it looks good! If you're reading this, good job WeatherSignal!

pressureNET: https://play.google.com/store/apps/details?id=ca.cumulonimbu...

Thanks very much! PressureNET's a really cool app as well.
I actually thought this post was about pressureNET at first :)
Kinda weird that the phone image shows a Galaxy S3 with physical back and home buttons, but then shows the software buttons...
Is this the intended use of the barometric sensors on phones like my Nexus 4?

Can anyone think of other uses?

I'm pretty sure the barometers are meant to be used for predicting height (i.e. as an altimeter) so that the phone can get a GPS lock more easily.
Interesting. Didn't know that. Thanks!
Yep, using pressure to gauge height is how many altimeters work. Then by removing this degree of freedom your phone is able to calculate a GPS location quicker as it knows you are confined to a particular area as opposed to a volume. I believe it also means you can get a GPS location when locked to one fewer satellites (though it may be slower in this situation).
One thing that I think would be amazing would be using gps and/or cell radio signals to do a very rough estimation of atmospheric conditions. Having done a bit of research with in geodesy at MIT, I remember using Trimble GPS receivers to track continental drift but having to account for environmental (mostly atmospheric) effects adding noise to our GPS readings. With a bit of low level signals processing, I believe it'd be possible to make plausible weather readings based on the noise in GPS receivers, though having done little mobile development, I'm not sure whether the Android OS would allow such low level access to GPS chips (or if the chips themselves are precise enough for the noise to come from the signal rather than the chip itself).
On the dashboard, there are four measures: Temperature, Light, Pressure, and Magnetic Flux. I know that my phone (Galaxy Nexus) has the last three, but I wonder where the temperature data is coming from. The only temperature data I can find with AndroSensor is the battery temperature. Are all these values live readings of my sensors?

[edit] I should note that the temperature on the dashboard is ~85 degrees Farenheit, when the current temperature is about 20F lower.

They're using the battery temperature run through an algorithm they've devised which they've found to, when averaged with other local users, estimate the actual temperature. My phone's been charging since I installed the app and that skyrockets the battery's temperature, so I have yet to see it give a remotely accurate reading of the ambient temperature.
Yes, this is how we do it. It's going to be quite rough for an individual user for an individual moment in time but we've shown that averaged over many users that battery temperature is closely correlated with the ambient environmental temperature. You should see greater accuracy when your phone has been relatively idle and reached an equilibrium with the environment whereas if your phone has been whirring through cpu cycles the accuracy will decrease.
Wish they'd just use the pressureNET data, since it's publicly available.
I don't think that their data is publicly available
We provide a livestream API to researchers only right now: http://pressurenet.cumulonimbus.ca/livestream/. We haven't yet opened to businesses yet because our new open source SDK is going to change our API quite a bit, so we want to stabilize it first. We're interested in this type of partnership however, and look forward to opening this up in the coming days/weeks. SDK code is about 90% feature complete and totally stable, so we're almost ready.

https://github.com/CbSoftware/pressureNET-SDK

(comment deleted)