Research On GB-InSAR Deformation Monitoring Technology Based On SFCW

In recent years, Ground Based Interferometric Synthetic Aperture Radar has attracted wide attention in the field of deformation monitoring, such as landslide monitoring, dam monitoring and glacier monitoring, etc. The system has high measurement precision, distant measurement and can contact or no contact measurement of the target in the range of a few meters to several kilometers and so on. For the process of GB-InSAR deformation monitoring, the main work done in the thesis is as follows:Firstly, the signal selection of GBSAR system is analyzed. From the perspective of the need of high resolution images in GBSAR system, Frequency Modulated Continuous Wave and Stepped Frequency Continuous Wave are selected. It is easy for the two signals to obtain large wideband. From the signal features, the signal echo model and imaging processing, etc., combined with requirements of GBSAR system in deformation monitoring, such as quick and accurate imaging algorithm, Stepped Frequency Continuous Wave is selected as the signal of GBSAR system. Then how the GBSAR system works is designed. Complex constraint relationships between various key parameters are analyzed and the design principles of the relevant parameters to be met are derived.Secondly, the two-dimensional echo model of GBSAR system based on frequency step and position step is built up. By studying the echo model, frequency domain back-propagation imaging algorithm based on coordinate transformation for GBSAR system is proposed. And the simulation experimental results of point target and plane target have been provided and analyzed.Finally, the GB-InSAR deformation monitoring process based on SFCW is analyzed. GBSAR image registration algorithm is presented. Round cycle mean filtering method based on coherent weighted is used for filtering interferograms generated by the GBSAR images. According to the coherence or amplitude, the pixels are selected from the filtered interferograms. Phase unwrapping based on the minimum cost flow in irregular network is used to deal with the selected pixels. In view of influencing the deformation measurement precision of phase unwrapping error, the least square method based on singular value decomposition is adopted to estimate phase. In order to reduce the influence of atmospheric phase on the deformation measurement accuracy, two-dimensional atmospheric phase screen model based on range and azimuth is established. Then the least square method is used to estimate atmospheric phase. At last the deformation values of the targets are extracted. Meanwhile, deformation monitoring simulation is performed on multiple targets. Under different threshold and noise variance, statistics about the deformation error are done.