Research On Parameter Estimation Methods Of Scattering Center With Anisotropy Characterization

With the development of theoretical research and improvement of radar signal processing technology,the parameter estimation of target anisotropic scattering center has oriented the development of radar signal processing.Features such as physical size and typical component structure of target,which reveal the intrinsic characteristics of objects,could be obtained by extracting and analyzing the information of scattering centers in radar echo signals.The target scattering center is approximately considered isotropic in most parameter estimation cases,nevertheless minor angle approximation could lead to a comparably higher error if the anisotropy characterization is neglected in micro-motion researching or wide-angle imaging.In this paper,an accurate completed scattering center is modeled based on electromagnetic scattering properties and analyzed the feature information of target by estimating the parameters of target anisotropic scattering center.The main contents of this paper are summarized as follows:1.The construction of the anisotropic scattering center model.This paper has introduced the conceptions of radar observation geometry and scattering center.The relational expression of measurements and scattering function is formulated based on radar observation geometry and scattering characteristics.A scattering center range profile model is derived from that model.Finally,the notion of anisotropic scattering center is introduced,whose model and representation are put forward and described in details.2.The researching of parameter estimation method based on sparse signal representation.We introduced the sparse representation of signal,regulation and inverse operation.We built a set of the aspect-dependent prior information based on object anisotropic characteristic,analyzed impact that the impulse waveform which was enlisted by the base vector had on the correlation of base vectors.Representing the anisotropic scattering center by matrix expression,we obtain a sparse over-completed matrix which included observation information.A sparsifying regularization method was proposed to solve the sparse representation problem of observation information which could be transforming to a non-linear optimization problem that we use quasi-Newton method to solve.Model parameter was then estimated to obtain the scattering center location and migratory radius.A 3D observation model was obtained by extending 2D counterpart.A orthogonal matching pursuit was brought up to estimate model parameter of 3D anisotropic scattering centers,and revise the parameter by using alternating projection.Proposed algorithm was proved performing favorably in experiment with data produced by CST studio.3.Research on scatting centers parameter estimation method base on high resolution range profiles.We described a weighted least squares-based iterative adaptive approach for amplitude and phase estimation(IAA-APES).The amplitude estimation and frequency estimation of this approach can both satisfy Cramer-Rao bound.Experimental results on large-scale datasets show that range profiles of the proposed algorithm have higher main lobe,lower side lobe and higher resolution.This paper also puts forward the 2D and 3D scattering center range profiles model separately and correlate amplitude and phase information of each scattering center based on target high resolution range profiles.Scattering center location and motion information is estimated by solving several equators.The approach has higher accuracy and lower compute burden by using cosine basis.Experimental results on large-scale datasets show that proposed approach can perform favorably.