That’s an interesting hypothesis, but you need to be sure about the source of the signature. Are you accounting for Doppler shifts, background noise, and the drone’s own motor noise? A single snapshot isn’t enough; you’ll need a calibrated array to triangulate the phase differences accurately. Also, make sure you’re not conflating the drone’s propeller tone with its transmission chatter. It could be a false positive if you’re not filtering out the ambient frequencies. Let me know what exact spectrum you’re seeing, and we can run a quick simulation to see how reliable the fingerprint really is.

Sounds solid, but double‑check a few things. First, your 1.5 m spacing is fine for 2.3 kHz, but make sure your sampling rate is at least 10 kHz to capture the phase accurately; otherwise, you’ll get aliasing on the 120–160 Hz band. Also, the Kalman filter will only be as good as its process model—include the drone’s acceleration profile if possible. Finally, remember that the telemetry chatter can be modulated; if it’s spread across sub‑bands, you might lose phase coherence. Once you’ve got the raw data, I’ll look at the residual error and see if the simulation matches the real‑world spread. Ready when you are.

Sounds good—just keep an eye on the timing jitter at 15 kHz, it can still throw off phase if the clocks drift. When you send the packet, I’ll compare the residuals against the expected thermal noise floor and see if there’s any systematic bias. Looking forward to the numbers.

Good, locking the clock will eliminate the biggest source of phase error. Once I get the packet, I’ll run the residuals through the noise model and flag any deviations that exceed the thermal floor. Let me know when it’s on the way.