Research On Radar Echo Reconstruction Technology Of Aerospace Target Under Sparse Conditions

Aerospace targets change rapidly in a short time,and the radar echo data obtained by a single radar are usually sparse in azimuth,which seriously affects the imaging quality and recognition rate of radar targets,and makes great troubles in comprehensive situational awareness.In order to obtain the overcomplete RCS data of radar targets,the traditional stepped frequency RCS measurement method in anechoic chambers usually needs to measure RCS data of radar targets at each stepped frequency point and azimuth attitude,which has the problems of long test cycle and high measurement cost.Radar echo reconstruction technology can solve the aforementioned problems to a certain extent.The radar echo reconstruction method based on interpolation algorithm is one of the common methods,and can work under the conditions of low noise interference and high signal sampling rate.However,this kind of method does not make use of the scattering characteristics contained in echo data,and its reconstruction accuracy is seriously restricted by noise interference and data sparsity.Another typical method for scattering parameter extraction and radar echo reconstruction is based on modern spectrum estimation technology,and it is better than the method based on interpolation algorithms.However,this kind of method can not deal with non-uniform and incomplete radar echo data(the sampling frequency is lower than Nyquist Shannon sampling rate).To solve these problems,this paper makes full use of scattering characteristics in high frequency region and sparse recovery methods,to accomplish sparse radar echo reconstruction.The main works in this paper can be summarized as follows:1.The one-dimensional geometric theory of diffraction(GTD)scattering model and the two-dimensional GTD scattering model are analyzed in detail at first.On this basis,the boundary conditions of radar echo reconstruction method based on interpolation algorithms are derived both in range and in azimuth.Simulation experiments validate the effectiveness and boundary conditions of the interpolation reconstruction method,and draw the conclusions that "under certain sparse conditions,the reconstruction performance of cubic spline interpolation algorithm is the best,cubic Hermite interpolation and linear interpolation algorithm are the second,and the nearest neighbor interpolation algorithm is the worst" and "radar echo reconstruction based on interpolation algorithms can be considered as a band limited spectrum recovery method".2.For the scene of radar echo missing in range,a method of scattering parameter extraction and radar echo reconstruction in range based on GTD scattering model and iterative weighted least squares(IRLS)algorithm(GTD-IRLS)is proposed in this paper.The introduction of an iterative regularization parameters is conducive to accelerating the sparse optimization solution process.By combining sparse reconstruction theory with the one-dimensional GTD scattering model,the method can extract scattering parameters and reconstruct the sparse radar echoes under the conditions of non-uniform and incomplete data.It is of some significance to solve the problem of scarcity of echo data and reduce the space-time cost of stepped frequency RCS measurement in anechoic chambers.Simulation data and electromagnetic calculation data are used to validate the effectiveness and boundary conditions of the method.3.For the scene of radar echo missing in azimuth,a radar echo reconstruction method in azimuth based on the undamped exponential(UE)scattering model and sparse iterative covariance estimation(SPICE)algorithm(UE-SPICE)is proposed.Firstly,under the constraint of small rotation angle,the conclusion that the two-dimensional GTD scattering model can be equivalent to multiple UE models in azimuth is deduced,making the decoupling of the two-dimensional GTD scattering model in range and in azimuth.On this basis,the sparse representation of radar echo in azimuth based on UE scattering model is established,and the sparse optimization is achieved by the SPICE algorithm.The method has the advantages of none selection of regularization parameters and good anti-noise performance.It is of some significance to accelerate the establishment of radar characteristic database and reduce the space-time overhead in anechoic chambers.Simulation data,electromagnetic calculation data and measurement data in anechoic chambers are used to verify the effectiveness and boundary conditions of the method.