Matrix-Completion-Based Sparse Baselines Tomographic SAR

The three-dimensional imaging system of Synthetic Aperture Radar(SAR)is an extension of conventional two-dimensional SAR imaging method.Three-dimensional SAR has not only all the advantages of traditional two-dimensional SAR,such as all-time,all-weather and high-resolution,but also avoided the ambiguity problems in two-dimensional imaging caused by projection from the three-dimension space to two-dimensional scene.Tomographic SAR(TomoSAR)can generate high resolution three-dimensional imagery by combining multiple SAR images acquired on flight paths slightly separated in the elevation direction.The technique of TomoSAR imaging because TomoSAR imaging system does not require complex flight trajectory controls,it also have a true threedimensional imaging capability.Matrix Completion is a low-rank matrix recovery algorithm developed in recent years.The theory demonstrate that if a low-rank matrix with missing data satisfy certain conditions,the matrix can be recovered by solving a convex optimization problem.In TomoSAR three-dimensional imaging system,the tomography resolution is affected by tomographic aperture length and number of baselins.Meanwhile,the scene satisfies low-rank property,we can use Matrix Completion theory to recover missing data of sparse baselines with fixed aperture length.This paper will first take CT technology in medicine as an example to introduce the tomographic theory,then according to the signal properties and imaging model of TomoSAR to analyze CT's application in three-dimensional SAR imaging.Then introduce Matrix Completion theory and study a method which achieve high-resolution imaging by using Matrix Completion to recover missing data in SAR echo signal matrix.Then the SAR image registration algorithm is studied in detail,introduce correlation coefficient method and maximum spectral method to achieve pixel and sub-pixel image registration,not only control computation amount,but also guarantee a high registration accuracy.Finally,analyze sparse baselines effect on tomographic resolution and introduce TomoSAR imaging method based on Fourier transform.A height-dimensional focus method with sparse baselines based on Matrix Completion is proposed.