Estimating the rotation rate of UAV propellers using pitch estimation techniques

Accurate estimation of propellers rotation rate is crucial for effective radar classification of drones and provides quite valuable information on its potentialities. This paper investigates several approaches to estimate this rotation rate, based on pitch estimation techniques, focusing on the nonlinear least squares (NLS) method, the harmonic Multiple Signal Classification (MUSIC) algorithm, and an autocorrelation function (ACF)-based approach. Through simulated analyses and experimental tests, the robustness of the three techniques is assessed. While all methods look promising, the ACF-based approach demonstrates great robustness against both simulated and experimental scenarios. Additionally, to assess the rotation direction, a radar system is introduced that operates with two displaced receivers and exploits the cross-correlation between the two received signals to estimate the direction.