Research On Sonar Detection Range Prediction Based On Database

Passive source location is an important research topic in the field of array signal processing.This subject has extremely wide applications in civilian and military fields.However,in some practical applications,far-field sources and near-field sources often exist at the same time.If the far-field source positioning method is used directly,the near-field source parameters cannot be estimated.On the other hand,if the near-field source positioning method is used directly,there will be problems,such as high computational complexity,difficulty in separating the mixed sources,and estimation errors.Aiming at the problem of localization of mixed far-field and near-field sources,this paper studies a localization algorithm for mixed near-field and far-field sources based on subspace difference.The main research contents can be summarized as follows:Firstly,the source location model is constructed.Then,the functional relationship between phase difference and incident angle and distance in the spatial observation signal steering vector is analyzed.Secondly,based on the near-field mixed source localization model,the statistic-based localization algorithm is studied,including 2-DMUSIC based on fourth-order accumulation,second-order oblique projection and fourth-order and second-order statistic mixing.The basic principle and implementation process are analyzed and elaborated,and the simulation are compared and verified.At the same time,in order to reduce the computational complexity,the polynomial root is used instead of the spectral peak search.Finally,the subspace difference,covariance difference and improved hybrid subspace difference is studied.Taking the root mean square error as the evaluation standard,the simulation results are compared from the actual observation data of algorithm computation complexity,array element number,signal to noise ratio,snapshot number and multi-source small angle difference.Through the research in this paper,it is basically efficient and accurate to locate the far-near-field multi-target hybrid source.