Research On Adaptive Enhancement Technology For Acoustic Signal Of Multi-rotor UAV

With the popularity of consumer-grade UAVs(Unmanned aerial vehicle,UAV),multi-rotor UAVs are widely used in material transportation,security,disaster relief and other fields because of their flexibility and simple control.However,the rapid popularization of multi-rotor UAVs has also brought many hidden dangers,such as spying on privacy,endangering public safety,smuggling drug carriers,leaking secrets of state and so on.Aiming at the lack of effective laws and regulations for low-altitude multi-rotor UAV,and the lack of appropriate means to supervise low-altitude UAV,at the same time,because acoustic detection is based on passive array acoustic detection,which has the characteristics of simple principle,low cost and easy realization,this paper researches on an adaptive acoustic signal enhancement technology for multi-rotor UAV.It provides a solution for further detection and reconnaissance of multi-rotor UAV acoustic signal,which has a good theoretical significance and high practical value.In this paper,the acoustic signal of multi-rotor UAV and the common noise interference are analyzed and compared,and the frequency domain characteristics of the multi-rotor UAV acoustic signal are obtained.On this basis,the adaptive enhancement processing framework and evaluation indicators of this paper are designed.Firstly,the adaptive line-spectrum enhancement processing is performed on the single-channel multi-rotor UAV acoustic signal,and the line spectrum detection method based on the double threshold method is used to detect the position and frequency of the line spectrum and calibrate.The data of line spectrum are retained and processed in the next step.Then,the time difference of arrival(TDOA)between channels is obtained by using the normalized least mean square(NLMS)algorithm,and then the orientation of the source target is calculated by geometric method.After obtaining the orientation information,the ISODATA clustering analysis algorithm is used to cluster the sound source orientation information,and the sound source type of each frame is predicted and marked.According to the labeling results,the multi-rotor UAV acoustic signal is cancelled by adaptive interference cancellation,and ultimately an ideal enhancement effect is obtained.The results of simulation experiments and field measurements show that the multi-rotor UAV acoustic signal adaptive enhancement algorithm used in this paper has good performance and provides a convenient and effective means for enhancing the acoustic signal of multi-rotor UAV.