Study On Frequency-domain Imaging Algorithms For Bistatic Forward-looking SAR

Bistatic forward-looking SAR(BFSAR) system not only has some advantages, such as its flexibilities, high anti-jamming and anti-interference ability, rich information about objects, but also can effectively solve the problem that squinted SAR can not image for the targets right ahead in the flight direction. However, the special geometric configuration of BFSAR has brought great difficulties to the actual imaging process. In this thesis, frequency-domain imaging algorithms for BFSAR will be studied and simulation software for BFSAR imaging will be designed and accomplished. The main study contents and innovations are summarized as follows:1. Translational invariant(TI) case and translational variant(TV) case for BFSAR are mainly introduced and analyzed. Then, the geometric configurations of parallel-flight mode for BFSAR and toward ground BFSAR are established, and their range historys are analyzed, respectively. At the same time, the process of received echo information is derived.2. By using the two-dimensional(2-D) spectrum of 2-D principle of stationary phase(2-D POSP), a novel Omega-K(WK) algorithm based on non-uniform FFT(NUFFT) for parallel-flight mode of TV-BFSAR is proposed. Using NUFFT can make full efficient use of the collected data in the spatial frequency domain, eliminate the range-azimuth coupling and correct the range cell migration(RCM). The simulation comparisons of the proposed algorithm with conventional algorithms demonstrate that the proposed algorithm not only can fully focus all the targets, but also has the best focusing performance.3. By using the 2-D spectrum based on method of series reversion(MSR), an improved extended azimuth nonlinear chirp scaling(EANLCS) algorithm for parallel-flight mode of TV-BFSAR is proposed. This method not only analyzes the higher order terms of azimuth frequency modulation, but also takes the spatial variances of quadratic frequency modulation rate and cubic terms into consideration. Besides, in order to derive the expression of the chirp scaling phase function coefficients, a fourth-order azimuth filter function is applied before taking the EANLCS operation. The simulation comparisons demonstrate the effectiveness and advantages of the proposed algorithm.4. An approximation expression of range history for toward ground BFSAR is figured, and Doppler properties of toward ground BFSAR is derived that the Doppler centroid and the Doppler modulation rate are related to the position of target. Then, by using the 2-D spectrum based on MSR, a range-Doppler(RD) algorithm for toward ground BFSAR is derived. Finally, a simulation is performed to validate the presented algorithm.5. Simulation software for BFSAR imaging is designed and accomplished. Firstly, the basic rules followed by the simulation software are introduced. Then, a scheme is studied for the system architecture design, based on object oriented design principle. Besides, on the basis of function, the modules are divided. Finally, Visual C++ is utilized to accomplish the simulation software, and the pictures of interface illustrate the accomplished simulation software for BFSAR imaging can realize the basic modules, such as echo simulation, imaging processing, performance estimation, and so on.