We consider the problem of the identification of describing functions for the reduced time domain modeling of interconnects. In particular we identify the rational approximation of regular parts, once delays are exactly extracted, with a combination of vector fitting algorithm with a post-processing procedure based on a non linear least square refinement step. This improves the accuracy of the approximation obtained, and leads directly to rank-1 residue matrices. The technique is applied to a couple of test cases and a comparative evaluation of performances is carried out.
AN IDENTIFICATION TECHNIQUE FOR MACROMODELING OF LONG INTERCONNECTS / DE MAGISTRIS, Massimiliano; L., DE TOMMASI. - STAMPA. - 1:(2005), pp. 185-188. (Intervento presentato al convegno 8th Workshop on Signal Propagation on Interconnects (SPI) 2005 tenutosi a Garmisch-Partenkirchen (Germany) nel 10-13 maggio 2005) [10.1109/SPI.2005.1500940].
AN IDENTIFICATION TECHNIQUE FOR MACROMODELING OF LONG INTERCONNECTS
DE MAGISTRIS, MASSIMILIANO;
2005
Abstract
We consider the problem of the identification of describing functions for the reduced time domain modeling of interconnects. In particular we identify the rational approximation of regular parts, once delays are exactly extracted, with a combination of vector fitting algorithm with a post-processing procedure based on a non linear least square refinement step. This improves the accuracy of the approximation obtained, and leads directly to rank-1 residue matrices. The technique is applied to a couple of test cases and a comparative evaluation of performances is carried out.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
