Macromodeling of digital I/O Buffers

Resumo

This paper addresses the state of the art of macromodels’ development for high speed digital output buffers for Signal Integrity (SI) analysis and simultaneous switching noise (SSN). An overview of the behavioral black-box techniques used by the Input/Output Buffer Information Specification (IBIS) and the previous nonlinear parametric models is presented. Then, the performance of a sigmoid basis functions’ artificial neural network is investigated to model the device behavior. The proposed alternative is based on the estimation of suitable mathematical equations reproducing the external behavior of the device via system identification theory.