New Selection Criteria

[MDofPerth]

What do you think of adding selection criteria on the Web Map to filter:
* By Manufacturer (or perhaps by range of MAC addresses).
* Exclude mobile devices (by excluding android and iPhone devices by either manufacturer or by ranges of MAC addresses).

The reason for asking to exclude mobile devices, if that is possible, is because many of the data points on the Web Map appear to be iPhone- or Android-supported WiFi Hotspots which are transient and mobile. Have a look at the location of bus-stops near universities for examples of dozens of points all clustered on top of each other and virtually all of them pocket hotspots.

[bja]

In the U.S. you can tell pretty quickly by the SSID - they almost always have names like "Verizon MiFi XXXX Secure" or similar. I had a similar thought when watching the client in operation today. Driving down the highway I saw a lot of transient hotspots that clearly came from nearby cars. This data isn't very useful for location purposes, so it should be at least marked as such.

[d0uglas]

Bump.. Laced all over my area (Connecticut) are wifi hotspots of cable companies named optimumwifi and xfinitywifi, and those make up over two percent (https://wigle.net/stats#ssidstats) of all reported hits by everyone. So it would be nice in the app to be able to filter out such things, or for Wigle to filter them out upon the user uploading in order to get cleaner data.

Another false-positive problem that's more complicated to solve is that on the highway especially or walking through a place like an airport people are driving with or carrying mobile wifi hotspots, but I have no idea how those could be excluded, perhaps they share MAC patterns.

Doug

[ithink314]

Bump.. Laced all over my area (Connecticut) are wifi hotspots of cable companies named optimumwifi and xfinitywifi, and those make up over two percent (https://wigle.net/stats#ssidstats) of all reported hits by everyone. So it would be nice in the app to be able to filter out such things, or for Wigle to filter them out upon the user uploading in order to get cleaner data.

Another false-positive problem that's more complicated to solve is that on the highway especially or walking through a place like an airport people are driving with or carrying mobile wifi hotspots, but I have no idea how those could be excluded, perhaps they share MAC patterns.

Doug

It's clearly not just Connecticut (attached). I agree filtering-out seems like a natural option, but this would still leave 98%, which is still a lot of wifi. The search is apparently intended to limit results to relatively small numbers of results, since it only gives 100 at a time in the list format, and detailed data is only shown on maps when zoomed far in.

I wouldn't call mobile wifi "false positive." Those wifi are out there. They do move around. And somebody is probably very interested in those movements.

Attachments

Screenshot_2017-12-19_14-26-42.png (178.31 KiB) Viewed 19037 times