| With the progress of science and technology,the information of ground objects obtained from spaceborne radar is becoming more and more accurate.The extraction of3 d information from the height information and the extraction of object deformation information are the important development direction of SAR application.With the development of synthetic aperture radar,from InSAR to multiple baseline synthetic aperture radar development,the information that SAR can obtain is also growing,and avoiding the interference in the 2d imaging can’t distinguish different scatterer within the same pixel scatterer.Considering the high cost of manual patrol,the inconvenience during the earthquake and and ice disaster natural conditions and personnel security,the multiple baseline synthetic aperture radar imaging has a broad application prospect.This article obtains from the principle of synthetic aperture,follow the 3d information inversion technology developing path,from the height of the In SAR introduce to the principle of information acquisition,in order to make up for the inadequacy of 2d imaging,combined with the principle of medical CT,SAR tomography process and application are expounded,and the multiple baseline chromatography model is established.In this paper,the compensation by singular value decomposition for non-uniform baselines and the estimation of the number of scattered bodies in a single pixel are studied.In the tomography algorithm of concrete,based on the spectral estimation algorithm efficiency is higher than traditional tomography algorithm based on Fourier transform,for the two specific spectrum estimation algorithm is studied,namely the multiple signal classification(MUSIC)algorithm and signal parameter estimation based on the technology of rotation invariant(ESPRIT)algorithm,although the spectrum estimation algorithm to obtain the results of spectrum for pseudo-spectral,still can be gained through its highly to the data information.Considering based on the compression perception theory can break through the Nyquist sampling theorem,and because the elevation of the radar to obtain information to meet the compressed sparsity,on the basis of perception to be below the Nyquist sampling rate by a small amount of samples to reconstruct accurate signal recovery.Studying SAR tomography algorithm based on compressed sensing theory,and compared with spectral estimation algorithm,in the simulation situation,studying thebaseline number influence on the height to the error information,uniform and non-uniform baseline conditions and the results of different algorithms signal-to-noise ratio on the tomographic results.Finally,the algorithm is brought into the actual data to deal with the scattering information of the tower,and the accuracy of different algorithms is compared with the actual height of the tower.The result of the SL1 MMER algorithm is the best in the case of low noise.It lays the foundation for the further study of the deformation information of the tower. |
