Study On Image Formation For Circular SAR And Related Technologies

Circular SAR is a new three-dimension SAR system which illuminates the regionwith circle path. In traditional SAR systems, the radar-carrying aircraft moves along aline which restricts the bandwidth of spatial frequency domain. Circular SARovercomes this restriction with circle path which expands the bandwidth of receiveddata, and in this way the resolution of image is improved. In addition, circular SAR canobtain the RCS information of whole aspect angle that is very useful for the detection ofconcealed target and special region that interested, so it is broad applied in military andcivilian. This thesis elaborates the principle of circular SAR and the image formationalgorithms. At last, several problems that have a strong impact on image quality areproposed to analysis. The structure of this thesis is illustrated as below:1. The principle of circular SAR. First the basic geometry model of circular SAR isintroduced in this thesis. Following the formula of echo signal is analysed in both timedomain and frequency domain. After that the three-dimension resolution of full apertureand partial aperture are proposed. Finally two traditional image formations of circularSAR are presented. One is Back Projection algorithm and the other is circular imageformation algorithm based on wavefront reconstruction.2. Circular SAR image formation algorithm based on Nonuniform Fast FourierTransform (NUFFT). Since in traditional polar format algorithm the interpolated gridwe chose focuses on a small range of frequencies relative to the amount of data that wereceived to work with, a circular SAR image formation algorithm based on NUFFT isproposed in this thesis. This proposed method utilizes all received data and applies aFast Fourier Transform on polar coordinates, thus it can improve the resolution of thecircular SAR imagery. In addition this proposed method is more efficiency thantraditional algorithms. The simulation results are given below to demonstrate theadvantages of this algorithm.3. An improved feature enhanced image formation for circular SAR. Since thecoherent angle of the reflectors in the scene could not provide persistence response infull aperture which has a negative effect on the resolution of the imagery, an improved feature enhanced image formation method is presented in this thesis. This algorithmrecovers the targe scene in the addition of sparse aperture, and to reduce the sidelobesthe penalty function of the optimization model is multiplied by the weights. In this way,the imagery with enhanced resolution and reduced sidelobes is obtained. Further morethe three-dimension image formation algorithm for multiple-pass circular SAR isdeduced and the simulation results are also presented below to state the effective of thisalgorithm.4. The coherent angle of reflectors in circular SAR. Since various viewing angleswill cause different scattering centers to be illuminated, the coherent angle of observedobject for the measured data of circular SAR is under analysis, and the result shows thatthe coherent angle of the reflectors in the scene could not provide persistence angle in360degrees. Then the resolution of circular could not reach the theoretical value.5. The motion compensation of circular SAR. Because of the complexity of thepath of circular SAR, the range and azimuth received data are serious tied up. Thismade the motion compensation problem more difficulty. To deal with this problem, anautofocus Back-Projection algorithm is used in this thesis. This algorithm compensatesthe random delay phase in the echo that realizes the focusing of imagery. Simulationresults are given to demonstrate the effective of this algorithm. Further more this thesisexpands this autofocus Back-Projection algorithm to three-dimension image formationin circular SAR. Compare with traditional BP algorithm, this proposed algorithm ismore effective and advantageous.