Passive Detection Method And System Design Of Drone Based On Soun

With the maturity of the drone industry,the rapid development of the drone industry has brought many benefits to society and has been widely used in aerial photography,surveying and mapping,agriculture,and other scenarios,but the laws and regulations related to drones have not kept up.The development of the drone industry has led to the frequent occurrence of drone "black flying" incidents around the world,posing a serious threat to national social security and personal privacy.The method based on sound detection UAV has the advantages of low cost and simple principle.Combined with the acoustic sensor array,the spatial location of the UAV can be further realized.The method based on sound detection UAV has gradually become a research focus of current low-altitude security.This paper analyzes the time-frequency domain characteristics of the four-rotor UAV acoustic signal.According to the characteristics of the UAV acoustic signal,the sound characteristics suitable for UAV detection are constructed,and the UAV acoustic signal classification model is established.To optimize the parameters of variational mode decomposition(VMD)for the fitness function,the error of time delay estimation is reduced,the positioning accuracy of the planar quaternary array is improved,and the UAV passive acoustic target detection system is built.The main research contents are as follows:(1)Analyze the sound signal of UAV in the time-frequency domain,analyze the influence of environmental noise on the sound signal of UAV,and study the preprocessing technology of sound signal and the characteristic parameters of sound in the time-frequency domain,which lays the foundation for the research in subsequent chapters.theoretical basis.(2)A UAV detection model based on Grey Wolf Algorithm(GWO)optimized Support Vector Machine(SVM)is proposed.According to the sound characteristics of UAVs,a new fusion feature parameter is constructed,which combines Mel cepstral coefficients(MFCC)and inverted Mel cepstral coefficients(IMFCC)to better characterize the UAV sound signals.Fisher criterion is used.Perform feature dimension reduction,reduce feature redundancy,improve system performance,optimize parameters in SVM through GWO,and establish a UAV sound classification model.The experimental results show that this method has significant advantages over traditional detection methods.(3)A generalized cross-correlation delay estimation algorithm based on improved VMD is proposed.Through the optimal selection of VMD parameters,the optimal combination of the number of modal components and the penalty parameters is determined,to avoid over-and under-decomposition of the acoustic signal,and reduce the delay error.Human-machine target positioning.Experiments show that the proposed method can effectively reduce the delay error,improve the positioning accuracy,and realize the position positioning of the UAV.(4)The UAV passive acoustic target detection system based on STM32 is built with MDK5 as the platform.The main modules of the system are acoustic signal acquisition module,data processing module,etc.,and the effectiveness of the system is verified through multiple sets of experiments.The experimental results show that the system can basically realize the detection and positioning of UAVs,which is of great significance to national and social security.