Dual Compressed Sensing Method In The Application Of Electromagnetic Scattering Problems By Method Of Moments

At present,the Method of Moments(MoM)is one of the most important methods in computational electromagnetics,and is widely used because of its accuracy.As a representative of the integral equation methods,MoM needs to iterate each incident angle in the calculation of a wide-angle incidence problem.The wide-angle method combination with compression sensing(CS)has been developed recently.Although the proposed CS methods have obvious advantages in computational efficiency,there are still some difficulties with the storage and computation of dense impedance matrices.This paper proposes a dual CS method for solving electromagnetic scattering problems.In the new scheme,the recovery algorithms are used two times to reconstruct the induced current.,Theoretical analysis and numerical results have shown that the new method can not only reduce the storage,but also reduce the computational complexity.The main work of the dissertation is as follows:Firstly,the theoretical framework of the MoM is studied and used to solving electromagnetic scattering problems of two-dimensional conductors and dielectric bodies Meanwhile,CS theory and sparse representation,the observation matrix and recovery algorithm,which are frequently used in the recovery algorithm are introduced and the orthogonal matching pursuit algorithm(OMP).Secondly,a fast solution model of compression sensing combined with the moment of method is established.The excitation matrix and the corresponding impedance matrix are sparsely transformed.By setting a reasonable threshold,the undetermined equation is constructed,the induction current is reconstructed by the recovery algorithm.The numerical examples are used to verify the effectiveness of the method.Thirdly,Aiming at the analysis of electromagnetic scattering characteristics of rotational symmetry targets,the bodies of revolution high-order moment of method of asymmetric wavelet transform is proposed.Compared with the traditional method of moments,it has obvious advantages in computational efficiency and lays a foundation for follow-up work.At the same time,through the sparseness of the discrete wavelet transform,the lines of the sparse post-excited vector and the impedance matrix are extracted to form an undetermined equation.A rapid analysis method for the electromagnetic scattering characteristics of the bodies of revolution is formed.Finally,a new method for bilateral sparse transformation is proposed under the compressed sensing framework.A model composed of dual CS to solve electromagnetic scattering problem.Compared with the single sparse transform CS method and the traditional method of moments,the new method reduces the memory footprint,reduces the computational complexity and improves the computation time.The feasibility of the new method is verified through experiments.