| With the capability of operating in all-weather and all-day,spaceborne synthetic aperture radar(SAR)and interferometric synthetic aperture radar(InSAR)show their significant importance in the battle and civil fields.Because of the "minimum antenna constraint",traditional single-channel SAR system cannot achieve high-resolution and wide-swath(HRWS)observation simultaneously.Multichannel SAR system is an effective tool to overcome above constraint and has been one new trend in the future remote sensing.With the development of multichannel SAR technique,multichannel InSAR is tending to be an attractive research.In this dissertation,some key techniques for multichannel SAR and InSAR data processing have been studied.The main content of the dissertation can be summarized as follows:(1)Imaging key techniques for multichannel SAR in azimuth In chapter 3,some key techniques in multichannel SAR in azimuth have been studied.Firstly,the echo model and channel error model of multichannel SAR in azimuth have been established.Then,to address the problem of channel error calibration,a novel phase bias estimation algorithm based on Doppler spectrum optimization is proposed in this chapter.By exploiting the fact that phase bias would cause Doppler spectrum broadened,the phase biases can be successfully estimated by maximizing a sharpness optimization of the Doppler spectrum.The effectiveness of the algorithm is validated by experimental result carried out on the simulated data and SAR data collected by an air-borne multi-channel system.Finally,the performance of the unambiguous reconstruction methods is compared,which include unambiguous method based on transfer function and unambiguous method based on array signal processing.(2)Imaging key techniques for multichannel SAR in elevation In chapter 4,the echo model of multichannel SAR in elevation is established.And two channel error estimation approaches for multichannel SAR in elevation are introduced.Due to the phase bias between channels of multichannel SAR in elevation,the quality of the SAR images would decrease after Digital Beam-Forming(DBF).In order to solve the above problem,we present a phase bias estimation algorithm for multichannel SAR in elevation.Firstly,the complex interferograms are obtained by conjugate multiplication of the focused SAR images acquired by the adjacent channels.Then,the differential interferograms are obtained by conjugate multiplication of the adjacent interferograms.Finally,the phase bias between each channel and the reference one are estimated by the maximum sharpness optimization.The proposed algorithm is validated through experimental result of the data acquired by automobile based multichannel SAR system.By analyzing the influence of the array error on the null location of the difference beam pattern,a channel error estimation approach based on difference beam pattern is presented in this chapter.At first,the difference beam pattern is constructed by the receive channel in elevation.And then,the channel phase error can be obtained by minimizing the energy in the null location.The effectiveness of the approach is validated by the SAR data acquired by automobile based multichannel SAR system.(3)Key techniques in Multichannel InSAR signal processing The chapter 5 can be divided into two parts.In the first part,the phase unwrapping techniques for HRWS InSAR signal processing is studied.As one of multi-channel SAR system,multi-baseline InSAR can overcome the problem of phase under-sampling in difficult terrain areas which is unavoidable in single-baseline InSAR.Multi-baseline phase filtering and phase unwrapping for multi-baseline InSAR signal processing is studied in the second part.In general,it is difficult considerably to handle a large scale phase unwrapping problem under the limited processing condition.In order to deal with phase unwrapping for multichannel HRWS InSAR data,a phase unwrapping method based on L1-norm tiling strategy is proposed in this chapter.Firstly,the residues are divided into different areas by the way of the positive and negative residues pre-connecting.The optimal branch-cuts and weights are obtained by optimizing all the possible branch-cuts between residues with opposite polarity.The weights will be used in the optimizing of the network transformed by residues.Then the Minimum Cost Flows(MCF)algorithm is taken to optimization of the network transformed by the residues.Finally,the optimal pair connecting is obtained from all the opposite polarity residues connected to the same residue.The method has an advantage of high efficiency and it can ensure the consistency between local and global solutions.Experiments based on data acquired by spaceborne SAR demonstrate the effectiveness of the proposed method.The method is very suitable for phase unwrapping in HRWS InSAR data processing.A phase noise filtering method based on the shortest unwrapped phase compensation for multi-baseline InSAR is proposed in the chapter.After compensating the terrain contours of each interferogram through the unwrapped phase acquired by the shortest length of baseline,a more stable phase distribution in the filtering local window is obtained.More phase samples taken from all the interferograms are used to improve the filtering performance.The effectiveness of the proposed method is verified by the simulated data and the real data acquired by an auto-mobile based multi-baseline InSAR system.A multi-baseline phase unwrapping method based on the unwrapped phase assisted is presented in this chapter.The advantages of unwrapping the phase acquired by the shortest baseline are taken by the method,which are stable processing and high efficiency.Based on the relation of unwrapped phase and the length of baseline,the terrain phase in the interferogram acquired by the long length is compensated.And then the phase fringes are few in the long baseline interferogram,resulting in an easy phase unwrapping.The simulated experimental result shows that the method can overcome the problem of discontinuities in the unwrapped phase effectively. |
PDF Full Text Request