The malicious use of small unmanned aerial vehicles (UAVs) necessitates the development of effective countermeasures against such threats. Counter-UAV systems encompass detection, classification, and neutralization. Detection and classification can be performed using visual, audio, radar, or radio frequency (RF) sensors. This paper proposes a straightforward UAV classification scheme based on analyzing RF transmissions according to fundamental parameters such as bandwidth, duration, and center frequency. The statistics of these parameters are expected to be unique, enabling differentiation between various UAV models. The paper outlines the methodology for analyzing received waveforms to estimate the aforementioned parameters and their distributions. Computer vision tools are employed for spectrogram processing. The proposed approach is validated on a large dataset containing waveforms from eight UAV models. Three types of statistics are evaluated, demonstrating that each analyzed UAV exhibits distinct transmission-related features.
Jarosław Magiera (Sun,) studied this question.