Prediction of Stochastic Eye Diagrams via IC Equivalents and Lagrange Polynomials

This paper addresses the prediction of eye diagrams in high-speed data links with the inclusion of manufacturing tolerances. The statistical assessment of the system performance is done via the combined application of accurate and efficient IC models and of the stochastic collocation method with Lagrange interpolating polynomials. Numerical results on the computation of the eye opening profile for a realistic PCB interconnect with the inclusion of the effects of parameters uncertainties conclude the paper.