Research On The Surface And Volume Scatterings And Their Applications

The electromagnetic scattering from randomly rough surfaces and volume scatterers have been an active research topic of in many areas, such as meteorology, oceanography, environment, military and so on. This dissertation made intensive research on random rough surface and volume electromagnetic wave scattering from analytical theory. Based on these two aspects of researches, we also paid attention to the problem on EM wave scattering from vegetation canopy on the rough surface.In terms of scattering from randomly rough surfaces, this dissertation carefully treated statistical distribution of slope or normal vector of surface, which were oversimplified by former analytical models, and proposed a statistical model in conjunction with the integral equation formalism for electromagnetic scattering from Gaussian rough surfaces with small to moderate heights and Gaussian power spectra where the statistical features of the surface slopes and the effect of shadowing were included. The proposed model introduced joint probability density function of surface normals to characterize their random distribution; furthermore, to simplify multiple fold integral involved in calculating ensemble averaging of scattering amplitudes, a decomposition scheme of slope covariance matrix was presented, which contained explicit physical meaning relating to correlation of slopes, and made possible the entire theory deduction and final simple analytical formulae for scattering coefficient computation. Then based on this proposed statistical method and the small slope approach, we also proposed a new two-scale model for wave scattering from a composite surface. For such model, it may capture the actual scattering mechanisms and lead to more accurate predictions, also may have promising applications for electromagnetic scattering from the ocean surface, whose entire roughness spectrum can be usually discomposed into small- and large-scale components.For volume scattering, the conventional T-matrix approach based on extended boundary condition method is reported to suffer from convergence problems for particles with extreme geometries represented by very large aspect ratios. This dissertation proposed a new iterative technique based on the T-matrix approach for the electromagnetic scattering by dielectric cylinders with arbitrary length. In this method, hypothetic surfaces were used to divide a cylinder into a cluster of N identical sub-cylinder, for each the T matrix can be directly calculated. Since any two neighboring subcylinder are touching via the division interface, the conventional multiscatterer equation method is not directly applicable. The coupling among sub-cylinder and boundary conditions at the interfaces were taken care of in our approach. Additionally, we extended this method to the problem of electromagnetic scattering from a cluster of parallel dielectric circular cylinders. Due to the division of the elongated cylinders, the intersection volume of the circumscribing spheres is strongly reduced. Therefore, the proposed method can be applied to the cases of when any pair of cylinders is close to each other. It should be noted that since no approximation is introduced in the procedure, this approach is thus more rigorous.In the field of electromagnetic scattering from vegetation, this dissertation made research on backscattering from the soybean, a typical species of short vegetation, and presented a model incorporating the benefits of branch model and phase and amplitude correction theory. In modeling inter-plant structure, the model introduced an antenna theory to take into consideration the quasi-periodic characteristics common in agricultural planting practice. Moreover, the proposed statistical integral equation model was adopted for direct scattering contribution of only randomly rough surfaces instead of traditional Kirchhoff approximation or small perturbation method; and in terms of the ground-scatter bounce scattering, conventionally used Fresnel reflection coefficient was modified to make full consideration of the impact of roughness of underlying surfaces. Finally, comparisons of model predictions with measuremental results validated this proposed model.