Change Detection Using Global Navigation Satellite System Based Bistatic Differential Interferometric Synthetic Aperture Radar

Global Navigation Satellite System based Bistatic Differential Interferometric Synthetic Aperture Radar(GNSS-Bi DIn SAR)is a novel radar system that consists of navigation satellites as illuminators of opportunity and a near-ground receiver.It can implement deformation monitoring through differential interferometry processing of repeat-pass SAR image sequences.Benefited from a large number of transmitters,wide beam,short revisit cycle and simple receiver side devices,this type of system has the characteristics of temporal and spatial continuous coverage of the transmitted signal,easy interference,flexible deployment and low cost,indicating its huge potential in the area of continuous deformation monitoring.During the practical application of navigation satellite Bi SAR system,the following problems need to be solved:(1)The navigation satellites move in the middle and high orbit,the angular velocity is very low,so in order to achieve a certain azimuth resolution,a long synthetic time is required,yielding a long time-frequency synchronization processing time,affecting the practical application.Deformation measurement processing requires a large number of image sequences,which increases the amount of data to be synchronized.At the same time,the interference signal in the actual environment will cause the position error of the range migration curve,thus affecting the time synchronization;(2)There are significant differences in the transmission power and signal bandwidth between the navigation satellites and the special remote sensing satellites,so the GNSSBi SAR image quality is poor,the image interpretation is more difficult,affecting the information acquisition ability;(3)Navigation satellites have no need for precise orbit repetition,and thus the orbital adjustment period is quite long,therefore,the spatial baseline formed during the orbit repetition under the influence of the perturbation force continues to increase over a long period of time.Since the interference phase is related to the product of the target elevation and the spatial baseline,the target elevation error will seriously affect the deformation estimation accuracy.There is a non-linear relationship between the interference phase and the target elevation,and the elevation estimation is more difficult,and the interference phase contains error sources like atmospheric delay which can influence the elevation and deformation estimation accuracy.In this paper,GNSS-Bi SAR is set as the research object.For the above difficulties,studies are conducted in three aspects: the fast time-frequency synchronization,image sequence fusion and deformation estimation,the innovative research results are as follows:(1)Research on Bi SAR Fast Time-Frequency Synchronization of Navigation Satellite under InterferenceTraditional synchronization method is inefficient and does not consider the influence of interference.Therefore,a fast synchronization algorithm based on slow-time jump pulse compression and a virtual complex signal is proposed.Firstly,the azimuth-down-sampled original range migration curve is obtained by the slow-time jump pulse compression of the direct signal.Then,to solve the problem that the curve position is affected by the interference signal,the theoretical slope history is fitted and the components above second-order coefficient are obtained and removed from the curve,yielding an approximate linear migration curve position sequence.A virtual complex signal is then constructed using the curve position as a complex signal phase.In this way,the linear and constant term estimation problem can be transformed into the frequency estimation problem of single frequency signal under the influence of noise.The slope of the curve,as well as the constant term,could be estimated by spectral analysis of the signal.After the removal of these items and positions statistic,points with high quality could be obtained,and these points are interpolated in the original curve,achieving the migration curve recovery under the interference;Finally,the positions are used as a reference to divide the echo data into blocks and through parallel processing,a fast time-frequency synchronization could be achieved.Experiments using Beidou-2 navigation satellite as the illuminators have been carried out,the echo data from the point target and the actual scene have been processed to verify the effectiveness of the method.(2)GNSS-Bi SAR Image Sequence Fusion MethodTo solve the problem of the low signal-to-noise ratio of the Bi SAR images,a Bi SAR image fusion method based on coherent processing is proposed.By using the fine coherence between the images of the adjacent repeat pass orbit,the coherent compensation phase information is obtained by the interference processing,phase filtering and the accumulation of adjacent phases to realize the estimation of the coherent compensation phase between any two repeat pass images,which can then be used to achieve the revisit image sequence coherence processing.The experimental data processing results show that the proposed method can improve the signal-to-noise ratio significantly.To tackle the problem that the capability of single-angle image information acquisition is limited,a Bi SAR multi-angle image fusion method based on angle domain is proposed.Based on the echo data obtained at different angles,the multi-angle image sequence is obtained by bistatic SAR imaging in the same scene.The image fusion is realized by superimposing the image amplitude.The results of the experimental data show that the method can improve the image quality,and can form the target continuous contour,which reduces the difficulty of image interpretation.(3)Deformation Estimation based on GNSS-Bi DIn SARTo solve the problem that the increasing spatial baseline affects the deformation estimation accuracy,considering the periodic perturbation characteristics of repeat-pass position,a deformation monitoring scheme using long spatial baseline interference sequence to implement height estimation,short spatial baseline interference sequence to implement deformation estimation,multi satellites switch to improve the time resolution has been proposed.In this paper,a multi-baseline interferometric SAR elevation estimation method based on approximate weighted least squares estimation is proposed to solve the problem of asymmetric bistatic SAR elevation estimation.Using the covariance matrix diagonal elements of the interferometry phases instead of the covariance to implement a rapid approximate estimation of the elevation,then the interference phase is differentiated at the approximate values to achieve the linearization of the interference equations,and the iterative realization of the weighted least-square estimation could be used to estimate the elevation.To solve the problem of the residual atmospheric phase and channel offset phase which affect the accuracy of the deformation estimation,a method based on generalized least-square estimation to jointly estimate the residual refractive index,channel phase offset and deformation is proposed.Repeat-pass Bi SAR experiments have been conducted to validate the proposed methods.During the research,the long-term experimental scheme design,data acquisition and processing have been carried out in different locations using the multi-channel experimental system based on navigation signals,and the above methods have been fully verified.