Geolocation by aeroplane sound Doppler shift

Despite this new one being a bit niche, it seems the concept is opening up the whole topic of "side-channel information hidden in audio streams"

External Quote:
You can actually do this with far greater accuracy using just church bells.

Can't quite believe it, but I tested it and it works.Planes need a timestamp and unbroken sound.

I tested 56 normal YouTube videos: 0 worked.Bells need a few seconds.

The bell ringers' database (Dove's Guide) logs every bell tower in England, 5,896 of them, all unique Hz, with the note of its biggest bell.

From a video with few seconds of the audio, the bells in a Derbyshire walk picked out the one church. I'm stunned.
 
It's not that difficult to copy it, but any hurdle added lessen the amount of time it'll get abused. And you don't want Sitrec being cited in the news as the tool used by a killer to locate his victim.

GM4AJK:
Security by obscurity. There will likely be a ton of tools to do this by the end of next week. The best @Mick West can do is hold off on releasing it in Sitrec for a while.

@RobertBillyBob Didn't even have to wait till the end of next week. Wednesday was long enough: DOPPLERGEIST

External Quote:
DOPPLERGEIST is a source-available Python toolkit for aircraft Doppler audio geolocation. It detects acoustic flyover patterns in recordings, matches them against historical ADS-B flight trajectories, and ranks possible recording locations.

The project provides automatic audio discovery, aircraft-data acquisition, geographic search, and a locally hosted browser interface. It is built for acoustic geolocation research, audio forensics, and studying location information carried by background sound.


 
The cat is out of the bag. Restricting stuff now is pointless.
That is a logically fallacious argument. That something is available elsewhere is in no way a positive argument for it to be made available here. At the moment there have been presented one cogent argument against it being made available here - the stalking one - and zero arguments for making it available. As a disinterested observer, one side seems to be making a stronger case.
 
I'll get it working better with known time and location data first. Then try some unknown cases. Right now it's failing where it should succeed, so that needs fixing.
Known time and location might be a good feature for sitrec, though? Just some kind of output that can say "matching aircraft corroborated by audio", as an additional piece of evidence supporting the analysis.
 
That is a logically fallacious argument. That something is available elsewhere is in no way a positive argument for it to be made available here. At the moment there have been presented one cogent argument against it being made available here - the stalking one - and zero arguments for making it available. As a disinterested observer, one side seems to be making a stronger case.
I'm assuming "disinterested observer" means you don't use the site tools (such as Sitrec). Then here's my counterargument since you seem to want one. As a Sitrec user, adding this functionality will definitely bring an advantage where it applies. To mitigate the stalker argument, Sitrec is an advanced tool, and since much simpler-to-use tools will be implemented elsewhere, I doubt Sitrec would be a stalker's weapon of choice. If anything, the only thing I see right now (today) is delaying the release of the functionality in Sitrec until those other tools become available.
 
It's simply a privacy issue. If someone did not want their home address made public, then we should generally avoid publishing it.

But it's something that happens here using other tools. We look up people's names in databases. We look at their social media. For example, I know Chris Bledsoe's address. In general, personal information should not be posted. Approximate locations, like a town, are okay.

I doubt Sitrec would be a stalker's weapon of choice. If anything, the only thing I see right now (today) is delaying the release of the functionality in Sitrec until those other tools become available.
https://github.com/1337isnot1337/DOPPLERGEIST is available, and I'm sure others will be. It's just too fun a project not to do.

It's not quite as amazing as some of the hype. My two recordings here were pretty inconclusive, with surprisingly little useful traffic. I had to extend the algorithm to fit harmonics, and it is currently being driven by leading-edge AI rather than manual picking. Works fine for the example in the OP.

I'll release it to senior menbers. first. Maybe just that. We'll see.
 
Back
Top