Research For Phase Unwrapping Algorithm Of Interferogram Based On Kalman Filtering

Interferometry Synthetic Aperture Radar is a kind of radar observation technology with the development of information technology.It has the advantages of high accuracy and large area,and can accurately obtain the digital elevation model of the earth surface.In In SAR technology,the acquisition of the surface elevation model requires multiple steps to process.Phase unwrapping is one of the steps,and its unwrapping accuracy has an important impact on the accurate establishment of the elevation model.Therefore,the paper mainly focuses on the key step in In SAR technology—interferogram phase unwrapping.The main research contents are as follows:(1)Representative traditional classical phase unwrapping algorithms,including branch cutting method,the quality map guide method,least square method,and nonlinear filtering algorithms in recent years,including unscented Kalman filtering algorithm,based on extended Kalman filter state variable dimension of volume and linear Kalman filtering algorithm principle is analyzed,and through the simulation and measured data experiments,several kinds of algorithms are studied and analyzed.(2)In order to solve the problem that the traditional phase unwrapping algorithm is difficult to accurately unwrap the low SNR interferogram,a phase unwrapping algorithm for unscented information filtering with fading factor adjustment is proposed.First,the model equation of the system is extended to the Taylor series second order;Secondly,the phase unwrapping is accomplished by combining the gradient estimation algorithm based on Amended Matrix Pencil Model and an efficient quality-guided strategy based on heap sort;Finally,the effectiveness of the algorithm is demonstrated experimentally.(3)The rank filter is applied to the phase unwrapping technique for the first time,and an adaptive rank filter phase unwrapping algorithm is proposed.First,the Amended Matrix Pencil Model gradient estimation algorithm is used to obtain the required phase gradient information;Secondly,the residual is used in the algorithm recursion process to adjust the gain matrix;Finally,the simulation and experimental results show that the algorithm can get more robust results.