Study On Acoustic Scattering Echo Characteristics And Characteristics Of Underwater Small-sized Targets

Underwater static small-size target is an important research target in the field of underwater acoustic signal processing,in which the hemispherical cylinder shell target is a typical spatial inhomogeneous target.It is of great significance to study the acoustic scattering characteristics and characteristics of targets for underwater active sonar detection.In this paper,the semispherical cylindrical shell target is taken as the research object,and the simulation modeling and the echo data measured in the pool experiment are studied by using COMSOL finite element software.The acoustic scattering components of the target are separated by using timefrequency analysis method and bright spot model,and then the characteristics of the separated signal are extracted,and the feature space is established based on the spectral characteristics.The separability effect is judged based on Euclidean distance and the feature space is enriched by Hilbert-Huang transform.An angle feature space model is proposed for experimental measurement data,which improves the separability between classes.The main work of this paper is as follows:Firstly,the background and basic theoretical knowledge of acoustic scattering from small underwater targets are introduced.The development of acoustic scattering research of underwater small-sized targets is briefly described from three aspects: acoustic scattering theory research,acoustic scattering signal characteristics and acoustic scattering feature extraction.The physical acoustic methods commonly used in the study of target acoustic scattering and the bright spot model of target echo are emphatically introduced.According to the sub-targets of the research target,the parameters are calculated and the angle-time delay distribution diagram of the experimental target is obtained.The experimental target is simulated by COMSOL finite element software and compared with the theoretical calculation results.Secondly,when the transmitted signal is LFM,the echo components of target acoustic scattering are superimposed in both time domain and frequency domain.Aiming at the superposition of echo components,the commonly used time-frequency analysis methods of acoustic scattering of underwater targets are introduced.Fractional Fourier transform is selected to separate the acoustic scattering components of the target acoustic scattering echo signal through the fractional domain under the best matching order of fractional order,and the separation effect is analyzed through simulation signals.The separated signals are subjected to WVD respectively,which eliminates the interference of cross terms among the components of the superimposed signal,and obtains the distribution of each separated component in time domain,frequency domain and time-frequency domain respectively.Thirdly,the method of spectrum feature extraction is introduced and the target acoustic scattering feature space is established,in which Hilbert marginal spectrum in Hilbert-Huang transform is introduced,and the extraction of marginal spectrum features enriches the feature space model.Judging the separability of geometric acoustic scattering and elastic acoustic scattering by Euclidean distance.The separation effect is verified by COMSOL simulation signal and experimental data.Aiming at the spatial inhomogeneity of experimental signals,a multi-angle feature space model is proposed,and the features are selected according to the classification effect of each feature under different angles.In the feature selection,it is found that Hilbert marginal spectral entropy has a relatively stable classification effect for the experimental data in this paper.The research proves that the angular feature space model can effectively improve the distance between classes and improve the separation effect between components.