I wondered what would be the applicability of this technique.
With ADS-B OUT it is trivial to configure somebody else’s aircraft reg, so it is perhaps to detect that. But that will be of use only if not talking to ATC (unless faking the reg in the radio calls too).
With the com radio you can even identify different pilots. I can do that without special equipment except a radio 
. Unfortunately we don’t know which signal properties they exploited because they just say they used a neural network for the signal analysis.
With digital signals you know always (given enough redundancy) the sender’s input signal and can construct an ideal output signal. The difference between that and the actual received signal would then be the starting point for the calculation of a fingerprint. That’s plain old signal processing but nowadays you apparently can only receive funding when you promise to use a neural network.
For voice signals the above method obviously won’t work but there are a lot of other angles to attack and these are not limited to the transmitter itself. Engine noise for example. Or interefence from avionics.
Surely that would work with the Com radio as well?
A group of academics reckon they’ve found a way to uniquely fingerprint aeroplanes’ Automatic Dependent Surveillance-Broadcast (ADS-B) tracking transmitters – though an aviation infosec boffin says more research is needed to verify the new technique.
In a paper titled “Real-World ADS-B signal recognition based on Radio Frequency Fingerprinting,” three Chinese researchers describe what they said was a method of identifying unique transmitters fitted to aircraft – regardless of what identity code the equipment is broadcasting.
Not really surprising. The basic idea has been used for many decades.
