| The electromagnetic (EM) modeling of the radar backscattering from three dimensional (3D) time-evolving sea surfaces and the composite model of the target above sea surface is always one of key and difficult problems of the industry, mainly in the following aspects:firstly, the geometric model of the sea surface is time-dependent and always electrically large in the electromagnetic scattering problem, the study on the backscattering from time-evolving sea surface has become a difficult problem in computational electromagnetics. Besides, for the case of a target underlying on the sea surface, the backscattering from the composite model includes the complex interactions between the target and the sea surface, how to model the electromagnetic coupling between the target and the sea surface is also a difficult issue. Finally, regardless of the characteristics study of sea clutter or the backscattering from the complex composite sea-target model, due to the randomness of the sea, it is often necessary to simulate a large number of samples to get the statistical properties, thus leading to huge computional costs and complexities.As stated in this logical order of from geometrical model of sea surface, to electromagnetic scattering simulation, then to characteristics study of the radar echo. This thesis has systematically studied the computational model of electromagnetic scatterings from the 3D dynamic sea surface and the composite scattering from target and sea surface, analysed Doppler spectra of radar echo and the distribution of sea clutter and chaos characteristics, and discussed the synthetic aperture radar (SAR) imaging problem of the target on the sea background. The main work of this paper is as follows:1. Aiming to solve the electromagnetic scattering problem of electrically large scale sea surface at large incident angle, multilevel fast multipole algorithm (MLFMA) which is known as an accurate numerical method in computational electromagnetics is introduced to solve this problem. By using this method, the effects to the backscattering of the sea surface induced by the variations of frequency, temperature and sea state are studied. By using a quasi-static approach, the time-evolving radar echo is acquired, then the Doppler spectra are computed for the high frequency surface wave radar as well as the breaking wave at X band.2. In order to get the electromagnetic backscattering from the composite scattering problem of the low flying target above the sea surface, a hybrid iteration method is applied. With the help of some acceleration methods, including: fast field approximate (FaFFA) technique, successive iteration technique, this hybrid method is more practical and efficient. The improved the hybrid method can greatly reduce the CPU time, then the time varying radar echo from low flying target above the time-evolving sea surface can be simulated by using this method. The Doppler spectra can also be studied.3. A framework for analyzing the backscattering from the sea surface is eatablished. The analysis work is through three aspects:Doppler spectra, the distribution of sea clutter and nonlinear chaos characteristics. With this framework, the sea clutter is simulated with the variation of frequency bands. It is found that Weibull distribution is more suitable to describe the HF radar echoes of sea clutter. The sea clutter has proved to be a chaos system for in the simulation experiments it has a positive maximum Lyapunov exponent.4. In the research of the SAR imaging problem, the overall scattering from the target and the rough sea background are simulated by using those mentioned approaches. The imaging scene is more practical, since the radar echo includes the contributions of interaction from sea background. Finally, the compressed sensing algorithm is applied to reduce the amount of data under the premise of not affecting the image quality in the study of reconstruction algorithm for SAR image. |
PDF Full Text Request