Fast Electromagnetic Simulation Of Microstrip Circuits And Antenna Arrays

The rigorous full-wave analysis of monolithic microwave integrated circuits (MMIC) and microstrip antennas has been made possible due to large increases in computing power and the development of a number of numerical techniques. Among the possible numerical methods applied to the analysis of this type of structures, the spatial-domain method of moment (MoM) in the mixed-potential integral-equation (MPIE) formulation is considered as one of the most powerful models for comprehensive treatment of various complex planar structures.In this thesis, two fast algorithms have been applied to reduce the memory storage and computational complexity of MoM. One is the fast Fourier transform (FFT) accelerated algorithm. This method can be traced back to the well-known conjugate gradient FFT (CG-FFT) method, which is restricted to the uniform discretization. To overcome this limitation, another fast algorithm, the adaptive integral method (AIM) is proposed, which retains the advantages of the CG-FFT method as well as the excellent modeling capability offered by the triangular elements. Both of these two algorithms reduce the memery storage to O(N) and computation complexity to O(N log N), where Ar is the number of unknowns. However, the FFT technique can't reduce the iteration number of the Krylov subspace iterative solvers, which is largely depends on the spectral properties of the integral operator or the matrices of discrete linear systems. Several preconditioning techniques have been used to reduce the condition number of the operator equations.With these techniques together, several fast integral equation solvers for microstrip structures are developed, which converge much faster than the conventional ones. Examples of circuits and antennas are presented to demonstrate the accuracy and efficiency of these solvers.