Study On High Resolution Imaging Algorithm For High Squint Airborne Sar System

Synthetic aperture radar (SAR) is used to generate high-resolution images of radar reflectivity of the scene of interest. It can provide specific information in all day and all weather. Therefore, it becomes an important remote sensing tool. High resolution is the most important feature which makes SAR distinguished from traditional radar, and higher resolution is the unremitting pursuit in SAR development. Ultra-high resolution SAR is an important research direction in recent years.SAR imaging algorithm is the core of SAR signal processing. A number of algorithms have been developed to form images, among which, the polar format algorithm(PFA) is the most popular one. Unfortunately, due to the wavefront curvature, the original PFA has limitations on the scene size and geometric fidelity. These limitations can be alleviated by wavefront curvature compensation technologies. However, all the proposed methodologies adopt approximation in the wavefront curvature compensation, and therefore, the modified PFA still suffers from scene size limitation. Imaging efficiency is another problem to be considered. The improved PFA based on chirp scaling(CS) is presented in this dissertation. This algorithm does not need interpolation operation which brings heavy computational burden. The dissertation is organized as follows.Chapter 1 is the introduction which outlines the history of the SAR technology and reviews its recent development. The motivation of this research is discussed.Chapter 2 studies the polar format algorithm. Firstly, the principle of PFA is briefly introduced. Then, it shows that the polar format transform can also be interpreted as a combination of a range and an azimuth scaling processing.Chapter 3 presents the PFA based on the principle of CS.A novel implementation of PFA using the principle of chirp scaling is discussed. The imaging precision of this algorithm is the same to the original one, but its real-time processing capability is obviously improved.Chapter 4 presents a detailed analysis of the wavefront curvature compensation. Since the differential range was approximated during the wavefront curvature processing in the traditional PFA, it could not meet the precision requirement. To solve this problem, a more accurate formula to represent the wavefront curvature is derived.Chapter 5 summarizes the dissertation.