Extracting R, L, G, C Parameters of Dispersive Planar Transmission Lines from Measured S-Parameters using a Genetic Algorithm

Jianmin Zhang
Marina Koledintseva, Missouri University of Science and Technology
James L. Drewniak, Missouri University of Science and Technology
G. Antonini
Antonio Orlandi

This document has been relocated to http://scholarsmine.mst.edu/ele_comeng_facwork/1618

There were 8 downloads as of 28 Jun 2016.


Signal integrity (SI) analysis of printed circuit boards for high-speed digital design requires information on the per-unit-length R, L, G, C parameters of the transmission lines. However, these are not always available when the property of the dielectric medium used in the board is unknown. A method to extract R, L, G, and C parameters from parallel-plate and strip transmission line geometries is proposed. It is based on measured scattering parameters and analytical modeling. A genetic algorithm (GA) is used to optimize the extraction by minimizing the frequency domain discrepancy between an objective function, which is the measured scattering matrix parameter, |S21|, and a GA model based on transmission line theory. The extracted R, L, G, and C parameters are then used in a SPICE model for simulation. Good agreement has been achieved in the reported results.