Passive circuit model parameter extraction using genetic algorithms

A significant advantage of multichip technology is the ability to embed passive components (i.e., capacitors, resistors, and inductors) directly into the substrate at low cost. The extraction of circuit model parameters for embedded passive components operating at high frequencies is crucial for both circuit design and performance characterization. In this paper, a method for optimizing the extraction of these parameters using genetic algorithms (GAs) is presented. To do so, a set of integrated passive structures are fabricated, and their scattering parameters are measured for a range of frequencies from 45 MHz to 5 GHz. Optimal equivalent circuit models for these structures are derived from the S-parameter measurements. Predicted S-parameters for the optimized equivalent circuit are then obtained from HSPICE. The difference between the measured and predicted S-parameters in the frequency range of interest is used as the measure of the accuracy of the optimization results. We then compare GA-based optimization to optimization using the Levenberg-Marquardt (LM) algorithm. It is determined that the accuracy of the parameter values obtained is improved using GAs.