An improved dem reconstruction method based on shortest sub-interval in MF-InSAR

Interferometric synthetic aperture radar (InSAR) can realize the elevation extraction of the target. Through increasing the degree of freedom (DOF) during the topography calculation, multi-frequency InSAR technology can further achieve high-accuracy elevation reconstruction in steep terrain. Although the maximum likelihood estimation (MLE) method is widely used in the signal processing of multi-frequency InSAR, its performance can be influenced by phase noise. This paper proposes a creative algorithm which combines the maximum likelihood elevation estimation with the shortest sub-interval seeking operator (SSIS-MLE). It statistically averages the elevation estimation results of the MLE with class histogram, so that efficiently excludes the error estimation. Simulation results verify that this method not only maintains the high calculation efficiency of traditional MLE method, but also guarantees the high accuracy of elevation estimation.