Studies On Spatial And Temporal Correlation And Source Detection At The Continental Shelf And Slope Area

The characteristic of the spatial and temporal correlation at the continental shelf and continental slope areas is rather complex.At the areas,the spatial and temporal correlation attenuates obviously,which affects the performance of the sonar system for remote target detection severely.Supported by the Natural Science Foundation of China(No.11274252,11534009)and the National Basic Research Program of China,this dissertation studies the characteristic of the spatial and temporal correlation at the continental shelf and continental slope areas and the problem of the target detection in the presence of the spatial correlation attenuation.In the dissertation,some new detection methods are proposed to reduce the performance degradation caused by the spatial correlation attenuation,which provide the theory and application foundation for improving the detection performance of the sonar system.The main contributions are as follows:1?The characteristic of the spatial and temporal correlation at the continental shelf and continental slope area is studied,which mainly includes the horizontal longitudinal correlation and the temporal correlation of the sound field.Firstly,the definition and the calculation method are introduced.Then,the horizontal longitudinal correlation coefficient and the temporal correlation coefficient are derived theoretically by using the normal mode theory.Through the simulations,effects of the waveguide environment and the experiment parameters on the spatial and temporal correlation are studied.Finally,the horizontal longitudinal correlation length and the variation range of the correlation length at the continental shelf and continental slope area in South China Sea are calculated by using experiment data.2?The effects of the spatial correlation attenuation on the performance of the target detection is studied,and the robust target detection method using subarray and its improved method are proposed.Firstly,the detection performances of some common detection methods in the presence of the spatial correlation attenuation are studied via theoretical derivation and simulation calculation.Then,to reduce the performance degradation of the sonar system using a large aperture array caused by the spatial correlation attenuation,the incoherent subspace reduced-rank subarray detection method is proposed and the theoretical detection performance of the proposed detection method is derived.On this basis the optimal subarray division method is proposed.Simulation calculations and sea trials demonstrate the effectiveness of the proposed method.Finally,the efficient incoherent subspace reduced-rank subarray detection method is proposed,with an aim of reducing the operation time.3?The detection methods based on spatial correlation improvement are proposed.Firstly,considering the problem of the spatial correlation attenuation caused by the interference among the normal modes,the warping transformation is used to extract the low order normal modes,which can improve the spatial correlation.On this basis,the detection method based on the normal mode extraction is proposed.Then,aiming at the problem of the spatial correlation attenuation caused by waveguide disturbance,the orthogonal matching pursuit algorithm is used to realize the sparse decomposition and the reconstruction of the received signals.Through the sparse decomposition and the reconstruction,the small spectrum components of the received signals are rejected,which improves the spatial correlation of the signals.On this basis,the detection method based on the sparse decomposition and the reconstruction is proposed.Finally,simulations demonstrate the effectiveness of the proposed methods.