Y'all haven't been using airplane mode and it shows

Suggestions for WiGLE/JiGLE/DiGLE

7 posts • Page 1 of 1
It's been a while since I've looked at the world map and it has become apparent that people have been using Wigle on their intercontinental flights. Neat to look at but doesn't seem have have any practical value. Is there a way to filter out networks based on elevation? I'm too lazy to look at the raw data but if Wigle processes elevation data, we can possibly filter out networks discovered over 10,000 feet where appropriate.
Image
We do a fair bit of filtering already; honestly a lot of the problems with points right now are issues with our cluster detection. We have a new implementation in testing, but haven't rolled it out.

Where you see points "smeared" on the map, look for the two major population centers the points lie between. We should probably "stick" the point to one of those places.

Fwiw, it looks far worse than it is the further zoomed out on the map you are. Pick a cluster and zoom in, as the display resolution increases, the points turn out to be sparse, and balanced on the knife edge of our old cluster detection system.

We're working on it, but we all have day jobs, and the clustering infra is such a big change that it's hard to address. As a stopgap, you can filter by date and QoS and discard low confidence points in the meantime.
It's funny, when I see the map and select it to show only those discovered by me, it shows dots pretty much all over the world, including in the US, Africa, Middle East, etc, where I have never even been close to.

I think there are more issues at play here, cos it's is supposedly showing me having detected things from 1000s of KM away.
Image
you've detected things that are *most commonly seen* far away from places you've been. You were the first to spot them, and saw them someplace you were, but their regular "homes" end up being in other places, and the preponderance of observations have snapped them to their current locations over time! This happens with any number of mobile WiFi hotspot types, as well as in the cases of moves and duplicated MAC addresses.
I should note that when things "smear" instead of snap, it means that the set of observations is too diverse or too evenly distributed for our current code to detect the best cluster(s) - you'll see these in between major metro areas at a low zoom level, but if you zoom in, you'll find that the clouds are actually quite sparse compared to urban density.
of course we *could* reassign discoverer credit when a point turns out to be *verifiably* and *currently* someplace fundamentally different from it's initial discovered location...

-a
Personally I find it interesting to see items I've discovered settle back into their rightful map-spot. Shows how we intersect in this big wide world.

I really look forward to seeing the Bluetooth output on maps...I can only imagine how dispersed all those devices I've crossed paths with at one airport or another are in real life.
Image

7 posts • Page 1 of 1

Return to “WiGLE Project Suggestions”

Who is online

Users browsing this forum: No registered users and 4 guests