LAUSR

This paper presents a detailed analysis of quadcopter noise emitted in different flight regimes and suggests an innovative approach to quadcopter sound detection. Currently, quadcopters have become extremely popular and are used for various applications. In some case scenarios, these applications are not legal; therefore, they represent a threat, and early detection of quadcopters becomes imperative. However, conventional detection with radars and infrared cameras does not provide satisfactory results, since some quadcopters have relatively small dimensions, and their motors have low heat emissions. Thus, this paper focuses on the noise emitted by quadcopters, which is connected to the changing revolutions per minute of the electric quadcopter motors. Thus, the paper connects the unique quadcopter features with the emitted sounds, which can be used for detection in different acoustic environments. The suggested algorithm uses signal processing with a nonlinear function, which results in new spectral lines that correspond to rotational speed differences between the front and rear motors. The analysis shows that the processed signal mixed with a traffic noise signal can be detected when the level reaches 3 dB lower than the level of the surrounding noise.