65 lines
3.4 KiB
Markdown
65 lines
3.4 KiB
Markdown
|
# Measuring Air Temperature With Phone Batteries
|
||
|
|
|
||
|
|
OpenSignal---a company responsible for mapping wireless signal
|
||
|
|
strength by gathering data using mobile device software---noticed [an
|
||
|
|
interest correlation between battery temperature on devices and air
|
||
|
|
temperature][0].
|
||
|
|
|
||
|
|
> Aggregating daily battery temperature readings to city level revealed a
|
||
|
|
> strong correlation with historic outdoor air temperature. With a
|
||
|
|
> mathematical transformation, the average battery temperature across a
|
||
|
|
> group of phones gives the outdoor air temperature.
|
||
|
|
|
||
|
|
[0]: http://opensignal.com/reports/battery-temperature-weather/
|
||
|
|
|
||
|
|
<!-- more -->
|
||
|
|
|
||
|
|
**Note:** Graph renderings on their website require proprietary JavaScript, but
|
||
|
|
the article does describe it in detail, so it is not necessary. In
|
||
|
|
particular, note that, from [their provided equation][0], their scaling factor
|
||
|
|
`m' implies that there is a smaller variance in battery temperature in the
|
||
|
|
graph than there is in the actual air temperature, but that there is still a
|
||
|
|
correlation.
|
||
|
|
|
||
|
|
This is an interesting find. The article further states that "[...] we have
|
||
|
|
one data point where the Android data is actually more reliable than the
|
||
|
|
traditional source."
|
||
|
|
|
||
|
|
Such data can be very useful in providing decentralized data, so long as
|
||
|
|
[issues of privacy][1] are addressed. Doing so is not terribly difficult,
|
||
|
|
but would have a number of factors. In particular, the user would need the
|
||
|
|
means to submit data anonymously, which could be done via software/networks
|
||
|
|
such as [Tor][2]. GPS location data is certainly a privacy issue when it is
|
||
|
|
tied to your mobile device, but fortunately, it's unneeded: you can trust
|
||
|
|
your users to let you know where they reside by either (a) opting into using
|
||
|
|
location services or (b) allowing them to specify a location or approximate
|
||
|
|
location of their choosing (approximations would be important since a user
|
||
|
|
may not wish to change their location manually while they travel, say, to
|
||
|
|
and from work). If enough devices submit data, then legitimate data would
|
||
|
|
drown out those who are trying to purposefully pollute the database. Such an
|
||
|
|
example can be seen with Bitcoin, in which networks will [reach a consensus
|
||
|
|
on correct blockchains][3] so long as "a majority of computing power is
|
||
|
|
controlled by nodes that are not cooperating to attack the network". Of
|
||
|
|
course, users would be able to pollute the network by sending false data as
|
||
|
|
it is, and the [data is already tarnished from various factors such as body
|
||
|
|
heat][0].
|
||
|
|
|
||
|
|
Of course, I do assume that mobile devices will contain temperature sensors
|
||
|
|
in the future; [some already do][4] (but I cannot encourage their use, as
|
||
|
|
they use [proprietary software][5]). However, this is still a clever hack (I
|
||
|
|
suppose that term is redundant). In my searching while writing this article,
|
||
|
|
I did notice [prior examples of ambient temperature readings using Android
|
||
|
|
software][6] ([proprietary][5]), but the software does not aggregate data
|
||
|
|
for purposes of determining weather patterns.
|
||
|
|
|
||
|
|
Finally, please do not download OpenSignal's app; it too is
|
||
|
|
[proprietary][5]; this discussion was purely from a conceptual standpoint
|
||
|
|
and does not endorse any software.
|
||
|
|
|
||
|
|
[1]: /2013/08/london-trashcan-spies
|
||
|
|
[2]: https://www.torproject.org/
|
||
|
|
[3]: http://en.wikipedia.org/wiki/Protocol_of_Bitcoin
|
||
|
|
[4]: http://stackoverflow.com/a/11628921
|
||
|
|
[5]: http://www.gnu.org/philosophy/free-sw.html
|
||
|
|
[6]: https://play.google.com/store/apps/details?id=androidesko.android.electronicthermometer&hl=en
|