Research On Doppler Parameters Estimation And Motion Compensation For Bistatic Forward Looking SAR

Since bistatic forward-looking SAR(BFSAR) can image the area in front of flight direction, it has attracted much attention in recent years. In SAR imaging, the Doppler centroid error leads to the decreasing signal-to-noise ratio and the offset image; the Doppler frequency rate error causes the image to be blurred; and the high-order Doppler parameter error brings the image distortion. In actual SAR flight, the aircraft platform is susceptible to air flow and other effects, which results in motion errors and thereby affects the focused quality. The thesis will focus on BFSAR Doppler parameters estimation and motion compensation. The main contents include:1. The geometry model of BFSAR is established, the point echo model is given and the properties of range history and Doppler parameters are analyzed. It is found that in BFSAR, the Doppler centroid is large, the Doppler frequency rate is small and the high-order Doppler parameters cannot be ignored. At the same time, the phase error effect on chirp signal is analyzed. It is also known that the first-order phase error shifts the center frequency, the quadratic phase error broadens the compressed signal and increases the side-lobe level, the third-order phase error leads to the asymmetry main-lobe.2. The Radon-based and range cell migration(RCM) correction function based Doppler ambiguity resolvers are studied. After the Radon transform is used to estimate the angle of range compressed data, the Doppler ambiguity number can be obtained through the relationship between the slope of echo signal and the Doppler ambiguity number. After constructing the RCM correction function and minimizing the entropy of the cost function, the Doppler ambiguity number can be estimated. The simulation results show that these two methods can accurately estimate the Doppler ambiguity number.3. A DCFT and Newton based Doppler parameters estimation for BFSAR is proposed. After the RCM is corrected by the cross-correlation technique, DCFT is used to estimate the initial value of Doppler parameters and the accurate Doppler parameters can be achieved by Newton method. The simulation results show that the proposed algorithm can obtain more accurate Doppler parameters than both the geometry-based method and DCFT method.4. A discrete polynomial phase transform(DFT) based Doppler parameters estimation for BFSAR is proposed. After the echo signal is compressed along range, RCM is corrected and azimuth signal is reformed, DPT is used to estimate Doppler parameters from high-order to low-order. The simulation results show that the algorithm can not only estimate the low-order Doppler parameters accurately, but the third-order Doppler parameter.5. The motion error model along track direction, yaw direction and pitch direction are established respectively. The effect of constant velocity error and linear velocity error on imaging is analyzed. The simulation verifies that the theoretical model and the formula derivation are correct. A motion compensation method based on Doppler parameters estimation is proposed. The azimuth compression function constructed by estimated Doppler parameters and a BFSAR imaging algorithm are combined to obtain the focused image. Finally, real data processing verifies the effectiveness of the proposed method.