Eye Diagram System Margining Surrogate-Based Optimization in a Server Silicon Validation Platform

Cargando...
Miniatura

Fecha

2017-10

Autores

Rangel-Patiño, Francisco E.
Chávez-Hurtado, José L.
Viveros-Wacher, Andrés
Rayas-Sánchez, José E.
Hakim, Nagib

Título de la revista

ISSN de la revista

Título del volumen

Editor

EuMA

Resumen

Descripción

Exhaustive enumeration methods for the physical layer (PHY) tuning of high-speed input/output (HSIO) links are prohibitive under current silicon server time-to-market (TTM) commitments. An alternative is to perform optimization on a highly accurate surrogate model. However, to increase the accuracy of the model, the number of lab measurements required to derive it also increases. In this paper, we analyze several surrogate modeling methods and design of experiments techniques to find the coarse model that is capable of approximating the real system behavior without requiring a large amount of actual measurements. We perform a direct optimization on the best coarse models found and verify the response by measuring the real system at the optimal coarse model solution.

Palabras clave

Equalization, Eye Diagram, High-speed Interconnects, HSIO, Kriging, Neural Network, Optimization, Polynomial, Post-silicon Validation, Receiver, Signal Integrity, Support Vector Machines, Surrogate Models, USB3, DoE

Citación

F. E. Rangel-Patiño, J. L. Chávez-Hurtado, A. Viveros-Wacher, J. E. Rayas-Sánchez, and N. Hakim, “Eye diagram system margining surrogate-based optimization in a server silicon validation platform,” in European Microwave Conf. (EuMC-2017), Nuremberg, Germany, Oct. 2017, pp. 540-543.