Research On High-accuracy InSAR Phase Unwrapping Technique Via Iteration

Interferometric Synthetic Aperture Radar(InSAR)can obtain the target elevation information besides the unique advantages of all-day and all-climate observation,and has being widely applied to a wide range of topographic measuring.The phase unwrapping plays an important role in InSAR data processing.Based on the advantages of the global high precision unwrapping algorithm,the four fast Fourier transforms(4-FFT)algorithm is studied and improved,and extended to the multi-baseline phase unwrapping.The multi-baseline high precision 4-FFT algorithm is proposed,combining with the simulation and experimental data analysis the algorithm performance in the end.The main work and innovation are as follows:First,the basic principle of interferometric synthetic aperture radar is described.Firstly,the height measurement basic principle of InSAR is introduced,and the InSAR data processing flow is given.Then,the basic principle of phase unwrapping is expounded and the traditional phase unwrapping algorithms and advantages and disadvantages are briefly introduced.So the study of high precision unwrapping algorithm is necessary.Finally,evaluation system of phase unwrapping algorithm is given,which provides the basis for the phase unwrapping algorithm performance.Second,four fast Fourier transforms(4-FFT)phase unwrapping algorithm is studied.The principle of 4-FFT algorithm is analysesed.We obtain the unwrapping phase by calculating the Laplacian of the relationship between real phase and wrapped phase and calculating the Laplacian of the unwrapped phase with knowledge of only the wrapped phase and solving for real phase.The global high precision 4-FFT algorithm is obtained.Then compared the 4-FFT algorithm with the global least squares algorithm in principle,the results show that the 4-FFT algorithm has higher unwrapping precision,and the simulation and the measured data also illustrate the high precision.Third,an iterative high precision 4-FFT phase unwrapping algorithm is proposed.The proposed algorithm can solve the low unwrapping precision even failure of the traditional 4-FFT phase unwrapping algorithm in low signal-to-noise ratio.The algorithm uses the traditional 4-FFT algorithm to obtain the initial unwrapping phase,and establishes the wrapped phase error formula with the wrapped phase.The real phase error obtained by 4-FFT phase unwrapping algorithm is added to the initial unwrapping phase iteratively.In this process,the phase error is getting smaller,and the unwrapping precision is getting higher.Finally,compared with the traditional 4-FFT and least squares algorithm in different signal-to-noise ratios,the simulation and measured data show that the proposed algorithm can achieve good unwrapping performance in low or high signal-to-noise ratio.Fourth,an iterative multi-baseline high precision 4-FFT phase unwrapping algorithm is proposed.The proposed algorithm can solve the low unwrapping precision and the phase overlap caused by terrain height mutation in single-baseline phase unwrapping.Based on the relationship between the baseline length and the interferometric phase,the 4-FFT algorithm is extended to the multi-baseline phase unwrapping by using the other baselines phase information,and the unwrapping precision is further improved by the error iterative compensation technique.The simulation and measured data show that the phase overlap is solved by the fusion of the solved baseline phase and other baselines phase.Compared with the high precision multi-baseline least squares algorithm,the proposed method has higher unwrapping precision.