The Adaptive Integral Method For 3d Electromagnetic Scattering Study

The research of electromagnetic scattering from complex targets has been being the focus for a long time for strong engineering application requirement. As an exact numerical method, the Method of Moments has been extensively applied to electromagnetic simulations. However, MoM is very difficult to be used for the analysis of electrically large problems, due to excessive memory requirement and CPU times. In this paper, Adaptive Integral Method (AIM) which is one of fast and efficient methods based on MoM, is studied. Compared to the conventional MoM, AIM reduces the memory requirement and computational complexity dramatically. The algorithm has a memory requirement and computational complexity proportional to less than O ( N1.5)and O ( N1. 5logN).Firstly, the high frequency methods, low frequency methods and hybrid methods used for analysis of scattering are reviewed in the present paper. The basic principle of MoM and RWG basic function with Garlerkin method are extensively studied. Several popular fast algorithms based on MoM are compared in the following.Secondly, the adaptive integral method is studied in detail. The two key techniques used in AIM are introduced in roundly. They are: fast and efficient solutions of auxiliary basis function, and FFT used to accelerate the computation of Toeplitz matrix and vector multiplication. Adaptive integral method has been used to the electromagnetic scattering analysis of 3D structures. Subsequently Different parameters in the algorithm which affect the accuracy and efficiency of the AIM are discussed. The numerical results show that the memory requirement and computational complexity of the algorithm are less than O ( N1.5)and O ( N1. 5logN).For electrically large problems, the condition number of impendence matrix is very large, and furthermore, the ununiformity of the modeling leads to an ill-conditioned matrix. In AIM solver, the convergence rate depends on spectral properties of the AIM matrix, and the near-zone threshold distance has a dramatic impact on the memory and CPU time. A poorly conditioned matrix occurs through AIM when a smaller near-zone...