Improved Omega-k Algorithm And Motion Compensation Research

Synthetic aperture radar(synthetic aperture radar, SAR), not affected by the weather and time,has the advantages of wide swath as well as wide band of surveying and mapping, and can be implemented on the ground hot spots for a long time observation, Therefore, in fields of civilian and military,SAR has achieved wide application. For SAR data processing, imaging and motion compensation is the key step of data processing, and has been the key point and difficulty of the research. Although the traditional time domain imaging operation can also produce a well effect for the image focusing, but it need a large amount of computation, so the operation efficiency is very low. Frequency domain processing algorithms such as RD algorithm and its improved CS algorithm solves the problem of operating efficiency, it is still difficult for the problem of large squint and wide aperture processing.This paper is devoted to the study of high squint SAR imaging and motion compensation, which firstly introduces the background and current research situation of synthetic aperture radar imaging, then studies the basic theory of SAR imaging, and then the commonly used imaging algorithm and motion compensation algorithm are introduced, in the main content part includes the following two aspects:1. This paper puts forward an improved Omega-K algorithm to solve the problem of high squint. As the influence of strabismus, supporting area of echo signal two-dimensional spectrum occur serious tilt, resulting in the difficulty of stolt interpolation. To address this problem, a improved two-dimensional spectral solved method is proposed. We first apply the principle of stationary phase derived the spectrum of the traditional Omega-K algorithm,and then gives the spectral improvement for high squint configurations, and based on the new 2-D spectrum, we improve those two steps of reference function multiplication and stolt interpolation, then we carry out the analyze of point simulaition in different squint of positive side,medium squint and high squint, and compare the focused effect of our improved algorithm, the traditional algorithm and the RD algorithm.2. This paper puts forward a method search optimization PGA algorithm based onChebyshev approximation and series inversion. In this method we improve the step of phase gradient estimation in the traditional PGA algorithm. For all distance units, we first give them a weight based on their energy and get the phase gradient center value. Then we give the concept of the phase gradient step value, and the image entropy is mapped as a function of the phase gradient step value,then we use the Chebyshev inequality to approximate the image entropy, and using the series inversion to obtain the step value which we set. Then we get a more accurate phase error from using the phase center values and the phase gradient step value. We then compensate the phase error and it not only reduces the number of iterations of traditional PGA algorithm, but also get a good focus effect fot the low-contrast image. treate the image that have a large mumber of isolated strong scattering points. Then the image that have a large mumber of isolated strong scattering points and low contrast images are analyzed, and we compare the number of iterations and focused effect of our algorithms and the classical algorithms.This work is surpported by national natural science foundation of China(No. 61173092), new century excellent talents to surpport plan(No.66ZY110) and shaanxi province science and technology research and development projects(No. 2013KJXX-64).