The Research On SAR Imaging Algorithm And Technology

Synthetic aperture radar (SAR) is placed in moving platform such as airplane and satellite, and can obtain image reflecting dispersion characteristic of the objects. Because SAR is free of weather influence, and works in both day and night, and it can greatly improve radar's capability of gaining information, it is widely applicated and well developed in both civilian and military fields in recent years.This thesis discusses the fundamental theory of planar SAR, analyzes planar SAR resolving ability in range and azimuth dimensions and the characteristic of Range Cell Migration (RCM). The detailed principle and realization of the classical SAR imaging algorithm (Range-Doppler (RD) algorithm) are studied in three cases, in which the antenna squint angle is zero, the range walk is revised in a small antenna squint angle, and the range walk and range curvature are adjusted in a small antenna squint angle. The SAR imaging with adjusting the range walk and range curvature is tested through the computer simulation of simulated data.We describe the relation between the Doppler parameter and imaging geometry. The peak detection method of cross-range power spectrum, correlative function method in slow time domain, and sign-doppler method are researched to estimate the Doppler centroid. Map deflexion algorithm (MD algorithm) is studied to estimate the Doppler frequency rate. The measured data is used to verify related methods and compare with their characteristics.This paper studies SAR raw data compression. At first the basis of SAR raw data compression the quantization theory is discussed, and the quantization threshold and the quantization output of the Gaussian data and the Rayleigh distribution data are calculated by program. Secondly Block Adaptive Quantization (BAQ) compression algorithm and Block Adaptive Amplitude Phase Quantization (BAAPQ) compression algorithm are described. Finally by comparing with the two SAR imaging results of the SAR raw data before compression and after compression, we can verify that the two compression algorithms are effective.