check out www.placelab.com
Anthony Townsend wrote:
a couple of questions... this is a very cool project, but i wonder
about its sustainability given the very rapidly changing base of
installed APs
1) how sensitive is PlaceLab to APs not showing up due to power off,
decommission, moved, etc. ? that is, any calculations of say how many
APs would have to be missing before accuracy falls off significantly?
We have done experiments to measure something similar. The results
are in a paper that is currently under submission. We measured how
the accuracy of Place Lab would change as we reduced the amount of
information that was available. We did this by dropping more and more
portions of the placelab database (which is different from, but has
the same effect as, turning off the corresponding APs). Our results
show that in an urban area like Seattle, a drop of as much as 30% of
the APs is still not enough to affect the average accuracy.
example - if for some reason 10% of the APs in the placelab database
are not transmitting, how does that affect average accuracy?
2) who does the initial survey? users? is there mechanism for
ensuring quality/accuracy/precision? how can users be expected to do
good surveys in urban environments where you need high gain GPS?
We expect users to do the initial surveys or war drives. These will
typically be a motivated set of users (not your average mom-and-pop).
That said, the accuracy of submitted data is an issue. We have a few
mechanisms in place already to catch data that is way off (either
because of hardware malfunction or due to malicious behavior). We are
still working on ways to identify data that is only slightly bad (say
displaced by a few meters from the true location). If you have any
ideas to contribute to this, we are open to discussion!
3) what happens when there are long delays between surveys? does
this affect precision of location determination significantly?
We do not know the answer to this yet. One of the things that has
been on our back burners is a long term evaluation (say 4-6 months) of
the evolution of WiFi coverage in a neighborhood, and using that to
figure out how often one would need to war-drive a neighborhood.
will wi-fi be around long enough for this to be a good long-run
solution? (not that this makes it not worth doing, but i'm curious
whether there's a long term plan)
The Place Lab architecture is largely agnostic about the specifics of
the beaconing technology used for location. In fact, a couple of
people on the project are currently working on supporting Bluetooth
and GSM beacons in Place Lab. So, even if 802.11-based WiFi isn't
around for ever, as long as some ubiquitous beaconing technology
exists that can uniquely identify each independent beacon, Place Lab
is still relevant.
Hope this answers your questions.
--Yatin
-------------------------------------------------------
This SF.Net email sponsored by Black Hat Briefings & Training.
Attend Black Hat Briefings & Training, Las Vegas July 24-29 - digital
self defense, top technical experts, no vendor pitches, unmatched
networking opportunities. Visit www.blackhat.com
_______________________________________________
Placelab-users mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/placelab-users
--
NYCwireless - http://www.nycwireless.net/
Un/Subscribe: http://lists.nycwireless.net/mailman/listinfo/nycwireless/
Archives: http://lists.nycwireless.net/pipermail/nycwireless/
