| As part of the continuing evolution of the monostatic synthetic aperture radar (SAR) system, the bistatic SAR greatly expands the application of the SAR because of its flexibility and the ability to detect the low radar cross-section object. Spaceborne-airborne bistatic SAR, as a kind of translational variant BiSAR, is one of the hotpoint in the field of SAR application, because it could not only be used to imaging, but also to revealing the scattering feature further, and improving the accuracy of the SAR image.This dissertation mainly researches imaging algorithms and motion compensation methods for translational variant SA-BiSAR. The main contributions of the dissertation are as follows:1. Introduce the fundamental theory of SA-BiSAR imaging: first describe and analyze the track of satellite, and propose the uniform coordinate for imaging. Based on these, some formulas for beam synchroniziation are deduced. Second introduce the theory and space resolution of SA-BiSAR imaging.2. An imaging algorithm for SA-BiSAR with parallel trajectory is proposed: First the range approximation for SA-BiSAR with parallel trajectory is deduced, some modifications are applicated to range-Doppler (RD) algorithm and well focused results are obtained, however there are some distortions. Second the courses of distortion are analysised, and a correction method named Inverse-Projection is proposed.3. An imaging algorithm for SA-BiSAR with non-parallel trajectory in time domain is proposed: first based on the SIFFT algorithm and the analyziation of space truncation error, the non-uninformly spaced correction SIFFT (NUSC-SIFFT) algorithm is proposed, which extends the imaging range and makes the traditional SIFFT algorithm suitable for SA-BiSAR. Second based on NUSC-SIFFT, imaging different parts along range direction and obtained well focused images of large scene.4. A 2-D spectrum imaging algorithm for SA-BiSAR based on translational variant space and time domain expansion is proposed. First, a new approximation of slant range history is obtained based on both space domain and slow time domain expansion. The expansion in space domain is applied to get the linear function of scatters'positions, which is beneficial to correct distortions. And slow time domain expansion makes it easy to use the principle of stationary phase (PSP) to obtain the 2-D spectrum expression. Second, based on the spectrum, the imaging algorithm is proposed. Because the imaging space is spanned by two vectors, which are translational variant and non-orthogonal for SA-BiSAR, so there will be some distortions, and traditional correction method is not valid any more. Then a new correction method based on eliminating coupled phase terms is proposed, and the imaging simulations show that the 2-D spectrum works well in the general configuration and the validity of correction method is proved.5. First discusses the effects of motion measurement errors (MMEs) on Bistatic SAR images. After the theoretical derivation, we find the main factors of causing SAR images to severely defocus. Then searching in the directions of P-AVME and P-RVME, according to the quality of images, and find out the MMEs which should be compensated. So a reasonable criterion should be used to evaluate the quality of SAR images. Compare with some other criterions, the criterion named minimum auto-correlation variance of the image is analyzed and proved useful by simulation. At last, the MMEs were estimated and compensated by searching method and BP algorithm with this criterion.6. The velocitie error will cause the defocusing of images by bring a second-order phase item. Then a compensation method based on parameters estimation is proposed, which uses the echo of different known scatterers to estimate the velocitie error and compensate the errors. The simulation results prove the validity of this method. |
PDF Full Text Request