Font Size: a A A

Research On High Squint SAR Imaging

Posted on:2017-05-19Degree:DoctorType:Dissertation
Country:ChinaCandidate:Y H LiFull Text:PDF
GTID:1108330503955253Subject:Target detection and recognition
Abstract/Summary:PDF Full Text Request
High squint synthetic aperture radar(SAR) can detect the target earlier than the broadside SAR. It improves the flexibility of SAR and the ability to survey the large area. Based on high squint SAR, the scattering properties of the target at the different angles can be obtained, which will increase the capacity for targets recognition. Hence, high squint SAR has been used in many fields, especially for the military applications. The range migration of high squint SAR is very large, which results in the serious range-azimuth coupling. Beside, due to the characteristics of the platform, the motion of high squint SAR is complicated and the accuracy of the inertial navigation system(INS) is low. Because of these difficulties, the conventional techniques are incapable for high squint SAR imaging. In order to improve the techniques for high squint SAR, the imaging algorithm, parameter estimation, motion compensation and geometric correction are discussed in detail in this dissertation. The main contents and contributions of this dissertation are given as follows.1. A novel imaging algorithm for high squint SAR is proposed. In order to reduce the range-azimuth coupling, the range walk correction is performed in the azimuth domain firstly. To accommodate the azimuth dependence of range migration, a block processing method is provided. The data are divided into blocks in the azimuth frequency domain and then are taken into the time domain where the azimuth-dependent range migration cell correction is performed. The method is based on the fact that the azimuth times corresponding to the same Doppler frequency are different for the different targets in the same range gate. For the purpose of compensating the azimuth dependence of azimuth modulation, a new azimuth nonlinear chirp scaling(ANCS) method is derived. Considering that the regions of support in the azimuth frequency domain are different for targets in the same range gate, the ANCS method equalizes the azimuth focusing parameters by multiplying a high-order phase in the azimuth frequency domain. The proposed algorithm is verified by real SAR data processing, and high-resolution images of high squint SAR are obtained.2. To estimate and compensate the azimuth dependence of focusing parameters, an extended multiple aperture mapdrift(MAM) and a high-order azimuth dependence compensation method(HADC) are proposed. Due to the low accuracy of INS, the focusing parameters(i.e. the quadratic, cubic and quartic coefficients of azimuth modulation) and their azimuth dependence are inaccurate. In order to get the accurate parameters, the extended MAM utilizes the cross-correlation in a sliding window to detect the azimuth-variant relative shifts between the two subimages. Based on the azimuth-variant relative shifts, the focusing parameters and their azimuth dependence are obtained and then are applied into the proposed imaging algorithm. Because the proposed imaging algorithm cannot accommodate the high-order azimuth dependence of focusing parameters, HADC takes small segments of the image to separate the targets and performs inverse Fourier transform to compensate the corresponding residual phase error of each target in the time domain. The effectiveness of the method in practical applications is proved by the real SAR data processing.3. To remove the geometric distortion caused by the error of radar altitude, three altitude estimation methods based on SAR data are proposed. The first method makes use of the relationship between the altitude and the range variation of the Doppler centroid. The Doppler centroid is fitted as function of the range, and the altitude is obtained from the coefficients of the function. Based on the sum and difference channels, the second method measures the elevation angles deviating from the beam center in each range gate, and then estimates the altitude from the range variation of these angles. The third method first records the SAR data at different grazing angles and estimates the range corresponding to the beam center. By establishing several equations of the grazing angle, the range and the altitude, the errors can be reduced, and the altitude is obtained by solving the equations. The feasibility and precision of the three estimation methods have been analyzed and proved.4. A new interpolation method for geometric correction in high squint SAR is proposed. First, the point spread function(PSF) in the slant image is deduced. The directions of sidelobes in the directions of range and azimuth are not perpendicular and the spectrum of PSF is skewed or even folded. The condition for the folded spectrum is given. Then, based on the spectrum of PSF, the new two-dimensional sinc interpolation kernel is provided. The new kernel has the same shape with the spectrum of PSF. By the new kernel, the right geometric correction is achieved. The results of real SAR data demonstrate the validity of the proposed method.
Keywords/Search Tags:high squint SAR, imaging algorithm, parameter estimation, geometric correction
PDF Full Text Request
Related items