Research On Imaging Of Staggered Synthetic Aperture Radar

Synthetic aperture radar(SAR)is a kind of microwave imaging system with high resolution,it can get images of the Earth's surface whenever the weather conditions and sunlight illumination are proper or not.However,the conventional Space-borne SAR cannot get images with high azimuth resolution and wide range swath.In order to solve this problem,some researchers proposed an innovative concept,named Staggered SAR based on the variable pulse repetition frequency(PRF).The Staggered SAR can achieve imaging with high resolution and wide swath at the expense of nonuniform sampling in azimuth direction and periodic pulse loss of the echo signal.When the data volume of radar system must be limited in a low level,the PRF shouldn't be set too high,which may result in the azimuth ambiguity in SAR imaging.In order to deal with the problem,this thesis studies one azimuth ambiguity removing method based on the change detection.Besides,another two methods for reconstruction without ambiguity also are proposed.The main work of this thesis includes:1.The missing-data distribution formula of Staggered SAR is deduced and then the construction method of SAR echo model is introduced under the assumption of non-stop.In allusion to the Staggered SAR imaging in practical condition,the characteristic of Staggered SAR with lowsample is analysed.Besides,the offset formula between azimuth ambiguity and true target is also deduced.2.For the azimuth ambiguity of Staggered SAR imaging with lowsample,an azimuth ambiguity removing method based on the change detection is studied.The method gets two range sub-spectra signals through the division of original range spectrum.Then the Principal Component Analysis(PCA)and k-means are used to detect the location of azimuth ambiguity based on the different offset of two sub-spectra signal.At last,the data about azimuth ambiguity can be removed from original image data.3.As the method based on the change detection can't remove the ambiguity close to the true target and can be applied in limited condition,an unambiguous recontruction method based on Iterative Shrinkage and Thresholding Algorithm(ISTA)is studied.This method needs to build the spare linear signal model of Staggered SAR and the measure matrix of scene.After the quantitative iterations of ISTA,the last reconstruction result of sparse target is obtained.4.As the recontruction method based on ISTA gets to convergence slowly,a recontruction method based on Learned Iterative Shrinkage and Thresholding Algorithm(LISTA)is studied.The network construction of this method is changed from the iteration of ISTA.The measure matrix of SAR is used as a parameter of LISTA network.After the pending data is sent to LISTA network,the last unambiguous reconstruction result of sparse target is obtained.