Resonator Characteristics Mode Of Microwave Devices Algorithm

The paper is started with the formulas of frequency-domain algorithm and the theory of FDTD, associated with the power–shift method which traditionally solves the eigenmode problem. But the traditional methods to get the results are too slow, and can't deal with the problem when there are many modes together, so an improved Chebyshev polynomial iteration is also used. It can not only get the single mode, but also make the speed quickly, what's more, the results are based on the algorithm, and we developed the eigenvalue module which can compute the frequencies and mode patterns of resonance modes in high power cavities. The algorithm is used in the software-CHIPIC. We test several cavities and get the results. Compared to the theory results and others which are gotten from other simulation software, we find all the results are reasonable that they show the accuracy and superiority of the method.This paper is mainly focused on the following two parts:1. In the eigenvalue module of 2-D edition, it lays emphasis on the problem of extracting the isolated eigen frequency in spectrum and the iteration accelerating. The method is following:(1) First of all, based on the boundary problem of HELMHOLTZ equation and Finite-Difference Technique, the algorithm calculates the field in the resonant cavity and disperses the HELMHOLTZ equation, as a result of the formula : Ax =λx.(2) Secondly, according to the EIGENVALUE of Matrix Theory and Applied Iterative Methods, the eigenvalue module adopts a numerical approach which allows the improved CHEBYSHEV polynomial iteration which based on the power method to extract the isolated eigenvalue in the spectrum.2. In the eigenvalue module of 3-D edition, the function which searches the minimal eigenvalue is achieved. The method is following:(1) It's similar to the 2-D edition, first of all, based on the boundary problem of HELMHOLTZ equation and Finite-Difference Technique, the algorithm calculates the fields in the resonant cavity and disperses the HELMHOLTZ equations, as a result of the formulas above. (2) Secondly, we can search the minimal eigenvalue directly with the iteration technique of power-shift method.3. Finally, we insert the eigenvalue module written by FORTRAN language into the CHIPIC which will have the function of eigenvalue analysis.In the end of this paper, we exam the capability of the 2-D edition eigenvalue module in three kinds of physical models, and the capability of the 3-D edition eigenvalue module is tested by two other kinds of physical models. The testing results show the accuracy and superiority of the method.