Access point positioning approach/algorithm

[carlos.pb]

First things first, as this is my first post in the forum, I'd like to introduce myself.

I'm a spanish student who's currently working on my master thesis in a topic related with wardriving and access point positioning. That's how I knew about Wigle. One of the things I'm trying to do now is to create a map with "as accurate as possible" positions for each access point. I've been reading a lot about this topic and I know it's not a trivial matter.

What I would like to know is how Wigle actually determines the location of and access point. A simple approach would be something like "the GPS position in which I received the strongest signal" while there are some other more sofisticated techniques like, for instance, triangulation. ¿How does Wigle do this and where within the source code is implemented?

I have already checked out the source code of the application and look for this but it's hard to find anything as there are too many classes. Also I have not been able to find any information anywhere.

I hope I've explained myself properly.

Thank you so much in advance.

Cheers from Madrid.

[icurnet]

What I would like to know is how Wigle actually determines the location of and access point. A simple approach would be something like "the GPS position in which I received the strongest signal" while there are some other more sofisticated techniques like, for instance, triangulation. ¿How does Wigle do this and where within the source code is implemented?

That's easy, it is stored in the file you upload to wigle.net (or that you do not upload to wigle). It is a text file, you can open it and see the gps coordinates if you have a gps unit functional when you capture wifi traffic.

[carlos.pb]

That's easy, it is stored in the file you upload to wigle.net (or that you do not upload to wigle). It is a text file, you can open it and see the gps coordinates if you have a gps unit functional when you capture wifi traffic.

The point is where do those gps coordinates come from. I mean, gps coordinates determines _your_ position, not the access point's. There should be some kind of policy/approach/algorithm in order to locate the ap.

Like I said before, the simplest approach would be to save the coordinate with the highest signal strength, but that doesn't work really well when mapping the access points, so I guess the approach should be somehow more complex.

Cheers.

[uhtu]

you must not have been trying too hard.

what we do is covered in the faq:
http://wigle.net/gps/gps/main/faq/

what you're asking about with "as accurate as possible" is a broader question. if you're trying to determine the physical location of a beacon, look into directional antennae and civil engineering survey techniques. if you're trying to determine the radio propagation and signal coverage from a point for a given beacon, you have to take a different approach.

all the filtering, location aggregation, and data manipulation we do is on the server side processing and not part of the open source clients.

[carlos.pb]

I'm sorry I missed the FAQ, they are clarifying. Thank you for your answer anyway.

What I'm trying to do is to determine the physical location of the access point. I've read about directional antennas but the idea is to use just a mobile device. I think I'll give weighted means a try and see how it works.

Thank you again.

Cheers.