Deep neural classifiers analyzing high-frequency blade micro-Doppler modulations to distinguish drones from birds and kites.
Rotating carbon fiber propellers introduce periodic Doppler modulation sidebands around the central target reflection. High-resolution spectrograms extract blade tip speeds (100 to 250 m/s), distinguishing quadcopters from flapping avian wings.
Secondary micro-vibration harmonics reveal drone structural mass and motor load, allowing air defense and traffic controllers to estimate whether an approaching drone carries heavy unauthorized payloads.