论文信息 - Non-Linear Statistical Modelling of GaAs FET Integrated Circuits Using Principal Component Analysis Abstract

Non-Linear Statistical Modelling of GaAs FET Integrated Circuits Using Principal Component Analysis Abstract

A statistical non-linear model of GaAs FET MMIC ’s which makes use of principal components to describe distance-dependent variations of technological parameters is presented.The proposed model is composed of a nominal non-linear model for the GaAs FET and a correlation matrix which accounts for correlation among parameters of the active devices on the MMIC.The correlation matrix is orthogonalised and reduced by using the Principal Component Analysis (PCA)technique. Capability to reproduce statistical distribution has been checked by comparing measured and modelled S-parameters in the 1-50 GHz frequency range by means of hypothesis tests of equivalence.

[1] J. Carroll,et al. FET statistical modeling using parameter orthogonalization , 1996 .

[2] Francesco Centurelli,et al. A Non-Linear Statistical Model for GaAs FET Integrated Circuits , 2002 .

[3] David A. Fisher,et al. Accurate design centering and yield prediction using the 'truth model' (FET model and MMIC design) , 1991, 1991 IEEE MTT-S International Microwave Symposium Digest.

[4] Kai Chang,et al. Nonlinear statistical modeling and yield estimation technique for use in Monte Carlo simulations [microwave devices and ICs] , 2000 .

[5] J. E. Purviance,et al. A linear statistical FET model using principal component analysis , 1989 .