Triangulation

[DarkStar]

I really like this feature of WiGLE, except when you only get a one sided view of an AP it can be really off.

--Chris ^^

[bobzilla]

Yes, it is true that things can get skewed, but in practice it seems to be working rather well. I'm sure we could come up with something a bit more accurate, but if all you have is a line of data you can't be sure which side of the line the data is on, so it'll never be totally accurate unless you get non-linear data, at which point the weighted average works better as well.

[DaClyde]

I'm noticing a lot of APs in my area are positioned much differently in JiGLE than they are depicted on Zhrodague's server and even on the various maps for the WWWD. Sometimes it's a difference of as much as 50-60 yards. When new locations are added to the database, are the AP locations being adjusted to account for the new data? If so, I recently built a new cantenna and can restumble those areas to get more precise location data....would that help in the re-positioning of the points, or would it be easier just to dump all my results and me re-submit them?

I've also noticed that if you do much zooming in and out, the points become misaligned with the map, and I have to close it and start over.

[uhtu]

the points are weighted averages. the more times a point is seen, the more "accurate" (or is that precise the mapping will become. this is indicated in the JiGLE client as QoS. you won't get anything specific (location-wise) from inaccurate radio gear like commercial gps and wifi. you can only get a general sense of past reachability in the general area.

the zoom issue is known, and in the TODO list, but should only be affected at the maximum zoom depth.