Research On Interference Feature And Extractive Technique Of Active Sonar Target In Shallow Water Environment

Active sonar is a core equipment for ocean detection and target recognition,with target feature extraction serving as a critical technology.However,as targets become stealthy and miniature,their scattering signals may be too weak to extract the target’s inherent features in the coupled acoustic field of shallow water environments and targets.To address this issue,this paper proposes a novel solution whereby the target and environmental acoustic field are treated as a whole.The target features in the environment are extracted by studying the influence of the target’s existence on the interference features of the acoustic field.This paper conducts research from three aspects:(1)theoretical and simulation research,(2)feature extraction,(3)acquisition and analysis of real measurement data.The theoretical research establishes a model of interference features of active sonar target scattering fields in a shallow water environment and analyzes the feature formation mechanism through simulation.Frist,it establishes a mathematical model of the active sonar target scattering acoustic field using the normal mode theory of environmental acoustic field and target scattering features.And the interference feature model of the active sonar target field was established.Then,the interference stripe features of the environmental acoustic field,the active sonar monostatic mode target field,and the active sonar bistatic mode target field were obtained through simulation.Finally,it conducts a theoretical analysis of target field interference features,providing a theoretical basis for the next step of feature extraction.Regarding feature extraction,this study separated interference fringes based on interference images,extracting features such as target scattering function,motion state,and waveguide invariant distribution.Firstly,to address interference fringe aliasing under active sonar bistatic mode,a method using two-dimensional Fourier transform was proposed to separate interference fringes.Then,grayscale value projection and image segmentation methods were introduced to extract target scattering function and target motion trajectory features from the interference fringe structure of the active sonar target field.Finally,the Radon transform method was applied to extract the waveguide invariant distribution in active sonar interference feature extraction.For data acquisition and analysis,lake experiments were carried out to obtain the interference fringe image of a spherical target and extract the features of the target in the environmental acoustic field,further validating the theoretical model and feature extraction method.Firstly,based on the results of theoretical research and simulation analysis,this study set up a test method and the test system,and proposed experimental scenarios and testing conditions for obtaining interference images.Subsequently,lake tests were conducted at the Moganshan Lake test site to obtain interference images of the environmental acoustic field and the target acoustic field under different scenes and working conditions.Finally,the feature extraction method proposed in this study was applied to extract the target’s scattering function,target motion characteristics,and waveguide invariant distribution features.The research results will provide theoretical and methodological support for target state analysis,tracking,and identification.