Reseach On The Performance And Processing Of Millimeter-wave In SAR

Interferometric Synthetic Aperture Radar(In SAR) is a kind of SAR technology system that obtains high accuracy digital elevation model(DEM) in all day and all weather conditions. Compared with the other bands, millimeter-wave In SAR has shorter wavelength and lower penetration, which has higher resolution and is more suitable for measuring the DEM. When the uncertainty of systems is high, the signal-to-noise ratios of images and interferograms are lower, and the performance of phase unwrapping and the accuracy of DEM degrade greatly. Thus, in the aspect of In SAR processing, how to obtain the high quality interferometric phase images is one of the key problems for millimeter-wave In SAR. To this end, this thesis researchs the performance and processing of millimeter-wave In SAR, the main works are as follows:(1) The echo model, the imaging principle of SAR and the elevation measurement principle of In SAR are introduced; we find that millimeter-wave In SAR can measure the high resolution elevation, however, which is more sensitive to antenna phase center trajectory. To get better imaging quality, the back-projection algorithm imaging combining with IMU data is applied for airborne millimeter-wave SAR imaging.(2) The effect of terrain on the interferogram is studied in detail. We find that the images have lower coherence, when the scattering point is far away from the imaging plane. In order to obtain better interferogram quality, the smallest projection height plane or curved surface projection should be chose.(3) The effect of the channel difference caused by different PSFs on the interferogram is studied, and we find that a fixed factor between the different channels has no effect in interferograms; when the difference of PSF amplitude is not a multiple relation in the two channels, it will significantly affect the quality of interferograms, and the larger difference is, the poorer image quality is. To avoid the sidelobe crosstalk of channels, the maximum sidelobe difference should be less than 25 ~ 30 d B.(4) The effect of platform stability on the interferogram is studied, we find that the low frequency error caused by the platform stability can be removed by the IMU compensation,but the effect of the high frequency error cannot be eliminated during imaging. For the purpose of avoiding the lower interferogram quality caused by the high frequency error, data accuracy of measurement should be limited to be less than 45 10-? m.(5) The interferogram filtering process is studied, and a method solving phase jump problem is proposed. Thus, filtering algorithm can be used in interference filter directly. Median filtering, mean filtering, KNN filtering and a variety of filtering algorithms are studied. Further more, a regularization phase filter algorithm considering phase jump problem is proposed, which can get a sound filtering effect.In a word, this thesis analyzes the factors influencing the quality of interferogram, and comes up with a solution of phase jump problem in the interferogram, which introduces some new ideas to improve the quality of interferogram.