Study Of Efficient Algorithm For The Solution Of Electromagnetic Scattering

With the rapid development of modern industrial technology,the analysis of efficient electromagnetic scattering characteristics of targets plays an important role in microwave imaging,chip simulation,radar system design and so on.However,the analysis of some large-scale,complex electromagnetic problems often cost a lot of computing resources and time.In order to improve the calculation efficiency and reduce the complexity of the problem,the research and development of efficient electromagnetic scattering algorithm becomes particularly critical.Based on the above background,the fast algorithm multilevel Green's function interpolation method(MLGFIM)based on the traditional method of moment is studied and applied to the scattering analysis of complex medium targets and the three-dimensional electromagnetic inverse scattering problem.The main contents and innovations of this paper are as follows:Firstly,the electromagnetic scattering problem is studied based on the mathematical principle method of the moment.The method of moments based on surface discretization is used to analyze the metal structure.For the electromagnetic structure including medium,the moment of method based on Poggio-Miller-Chang-Harrington-Wu-Tsai(PMCHWT)equation and volume integral equation is used to adopted.In order to improve the efficiency of electromagnetic scattering calculation,we apply MLGFIM to three-dimensional full wave electromagnetic scattering analysis,and compare the accuracy and efficiency of traditional method and the fast method.Then,MLGFIM is firstly applied to solve the large-scale inverse scattering problem.In the inversion algorithm distorted born iteration method(DBIM),the repeated calculation of the electromagnetic scattering forward problem consumes a lot of time and memory,especially in the three-dimensional inverse scattering problem.An improved Bonn iterative inversion method is proposed.In order to make the inversion algorithm more stable,the truncated singular value decomposition(TSVD)technique is used to avoid the tedious selection of regularization terms.In order to further reduce the calculation,the total field in this function is replaced by the incident field caused by the updated inhomogeneous background medium.The whole algorithm flow is described in detail,the speed of inversion image algorithm has been improved obviously,and the image quality has been well guaranteed.