Research On Three-Dimensional Passive Source Localization Based On Sparse Decomposition Theory

Source localization is one of the long-standing research hotspots in the field of underwater acoustic signal processing.It is self-evident for China's influence on marine strategy and national defense construction.Compared with active positioning,the passive positioning detection distance is farther and the concealment is higher,so that in the underwater acoustic confrontation can increase the safety.Therefore,Passive positioning has important practical significance.In view of this,this paper will study the three-dimensional passive positioning of sources based on sparse decomposition theory,and seek to achieve high-resolution estimation of three spatial dimensional position parameters(azimuth,distance and depth)of the source of interest.According to the sparsity characteristics of the target distributed in the space domain,the problem of source localization can be transformed into an underdetermined linear system by means of sparse decomposition theory,and the appropriate sparse decomposition algorithm is used to solve the problem.In order to realize the three-dimensional positioning of the source,the main work of this paper can be divided into two parts,namely,the Direction of Arrival(DOA)and the Matched-Field Processing(MFP).The former realizes the source orientation.The latter is used to estimate the distance and depth of the source.Thereby achieving the purpose of three-dimensional positioning.The process of spatial spectral estimation and matched-field processing is similar.Firstly,according to the positioning requirements of the source position parameters,the corresponding search range is delineated,and the appropriate search step size is selected according to the positioning accuracy requirements,thereby forming a corresponding space search grid.Then,according to the different types of sources(narrowband,wideband,coherent,incoherent,etc.),the corresponding mathematical model of array receiving data is established based on the sparse decomposition theory.Finally,the sparse decomposition algorithm is used to solve the problem,and the positioning estimation result is obtained.We assume that there are each sources on all grid points,but only the grid points where the real source is located have a certain signal strength,and the rest are all zero.Therefore,the result obtained by the sparse decomposition algorithm is a vector or matrix of the same dimension as the search grid.According to the above process,the paper studies the spatial spectrum estimation and matched-field processing separately.The corresponding solutions are given for different sources and the number of received data snapshots.The simulation and experimental data processing results verify the validity and accuracy of these methods?...