Systematic Extraction of Static Images From Layered Media Green's Function for Accurate DCIM Implementation

The priority-queue-based ray tracing technique is utilized for convergence acceleration in the Michalski-Zhang's mixed potential dyadic Green's function components of multilayered media. The systematic nature of the prioritized ray tracing allows to determine the maximum number of extracted images beyond which the extraction does not improve the spectrum convergence any further, thus, yielding the fastest possible convergence in Sommerfeld integrands achievable through the closed-form static image extraction. Such information is shown to yield a substantial accuracy improvement in the Discrete Complex Image Method (DCIM)

I. Introduction

The efficient electromagnetic analysis of the interconnect structures requires implementation of the method of moments (MoM) in conjunction with the layered media Green's function [1]. Various techniques has been proposed in the past for the efficient treatment of such Green's functions [2], [3]. The effectiveness of these methods largely depends on the convergence rate of the integrands remaining after the analytic extraction of the quasi-static and surface wave contributions. While in many layered-media problems simple extraction of a fixed number of quasi-static images may suffice [4], today's semiconductor process technology calls for more elaborate extraction techniques where the number of extracted images can be adaptively varied to ensure a robust control over the convergence rate of the Sommerfeld integrands.