基于FRFT和PGA的混合自聚焦算法.docVIP

  A hybrid SAR autofocus algorithm based FRFT and PGA# WANG Dalong, BAI Xia** (Information and Electronics School,Beijing Institute of Technology, Beijing 100081) 5 10 15 20 25 30 35 Abstract: In order to obtain high-resolution images, autofocus algorithms are widely used in SAR imaging. In the paper, a hybrid SAR autofocus algorithm is proposed. The proposed method estimates the quadratic phase error by FRFT method and the residual phase error by PGA. Results from simulated and real data prove that performance of the proposed algorithm is superior to the individual PGA. Key words: synthetic aperture radar (SAR); autofocus; phase gradient autofocus (PGA); fractional Fourier transform (FRFT) 0 Introduction Synthetic aperture radar (SAR) is a kind of high resolution coherent imaging radar. To obtain a high resolution SAR image, accurate information of the relative motion between the radar and the target are required. So some navigational devices such as global positioning system (GPS) and inertial navigation systems (INS) are equipped on the SAR platform to provide information about the platform motion. Unfortunately the information provided by GPS and INS are not so precise that the uncompensated movement of the platform will cause phase turbulence, which results in a defocused image along cross-range direction. In order to compensate the phase error, a technique called autofocus arises, which estimates the motion parameters directly from the data. According to the frequency variation in a synthetic aperture time, the phase errors can be divided into high-frequency phase error and low-frequency phase error [1]. The high-frequency phase error usually compensated by GPS and IMU and the low-frequency phase error must be compensated by autofocus. Many methods of autofocus have been provided in literatures which can be divided as parametric method such as Contrast Optimization (CO), Map Drift algorithm (MD) [2], Shift and Correlation (SAC), and non-parametric method