3 ms·

I think you're mistaken about a few things here? 1: The network's used to triangulate that are tied to a GPS location retrieved from some historical log of whe

by cududa 9y ago

I think you're mistaken about a few things here? 1: The network's used to triangulate that are tied to a GPS location retrieved from some historical log of when a car drove by it and logged it as a known location. https://www.quora.com/How-does-WiFi-improve-the-GPS-location https://www.quora.com/How-does-WiFi-improve-the-GPS-location 2: How would boats on the open sea receive wifi?

exikyut 9y ago
OK, both fair points. I'm referring to https://youtu.be/dfLE_nXh7jY https://youtu.be/dfLE_nXh7jY, the video mentioned just after halfway down. Google Maps is also referred to elsewhere in the article as well, as are many anecdotes of people proving this with phones. Regarding #1, yes, Wi-Fi positioning is based on an initial "known good" data collection that is then improved by a quorum of unrelated devices continuously and consistently reporting a given group of SSIDs when at a certain GPS location. But if you have a device reporting, say, 10 or 20 hotspots, and that collection of however many SSIDs is unique in the entire world, and then the device GPS kicks in and says it's 20 or 200 or 2000 miles away from that location... and then a number of devices start reporting the same information... that's going to show up on a dashboard somewhere, because exactly the same kind of behavior would occur in a Wi-Fi positioning service attack situation. So the recipients of this conflicting information have a decision to make about whether to silently not report the conflict, or not. I wouldn't believe a line like "the algorithms are forgiving and will trust the same information reported from 50 devices as accurate"; a moderately determined actor could spin up hundreds of "devices" just by using SDK emulators, and anyone even more determined could figure out the Wi-Fi positioning protocol (I remember an article on here noting how Apple Wi-Fi positioning works - IIRC there was very little validation if you just wanted to find the location of a given list of SSIDs) and generate theoretically infinite "GPS/SSID correlation update" packets. Regarding #2, you are right; ships (particularly without VTS data service) rely entirely on GPS. I'm not challenging that though. If I have misunderstood your meaning please let me know. (No offense intended.)

by cududa 9y ago

exikyut 9y ago

OK, both fair points. I'm referring to https://youtu.be/dfLE_nXh7jY https://youtu.be/dfLE_nXh7jY, the video mentioned just after halfway down. Google Maps is also referred to elsewhere in the article as well, as are many anecdotes of people proving this with phones. Regarding #1, yes, Wi-Fi positioning is based on an initial "known good" data collection that is then improved by a quorum of unrelated devices continuously and consistently reporting a given group of SSIDs when at a certain GPS location. But if you have a device reporting, say, 10 or 20 hotspots, and that collection of however many SSIDs is unique in the entire world, and then the device GPS kicks in and says it's 20 or 200 or 2000 miles away from that location... and then a number of devices start reporting the same information... that's going to show up on a dashboard somewhere, because exactly the same kind of behavior would occur in a Wi-Fi positioning service attack situation. So the recipients of this conflicting information have a decision to make about whether to silently not report the conflict, or not. I wouldn't believe a line like "the algorithms are forgiving and will trust the same information reported from 50 devices as accurate"; a moderately determined actor could spin up hundreds of "devices" just by using SDK emulators, and anyone even more determined could figure out the Wi-Fi positioning protocol (I remember an article on here noting how Apple Wi-Fi positioning works - IIRC there was very little validation if you just wanted to find the location of a given list of SSIDs) and generate theoretically infinite "GPS/SSID correlation update" packets. Regarding #2, you are right; ships (particularly without VTS data service) rely entirely on GPS. I'm not challenging that though. If I have misunderstood your meaning please let me know. (No offense intended.)