On the Correlations Between Model Process Parameters in Statistical Modeling

J. Slezák, A. Litschmann, S. Banas, R. Mlcousek and M. Kejhar
ON Semiconductor, CZ

Keywords: Statistical modeling, statistical correlations, corner analysis, process variations, Monte Carlo

Abstract:
Statistical modeling in the design of todays high performance integrated circuits (ICs) is a necessity to produce competitive products with short development time. The use of backward propagation of variance (BPV) has proven its worth among other approaches proposed for the statistical modeling. This methodology introduces physically based process and geometry-dependent parameters (PGPs) for each device that is available to the design community in a model library. The goal of this contribution is to propose a simple method to establish mathematical relationships among correlated PGPs.

Nanotech 2004 Conference Technical Program Abstract