Artificial neural network to statistically model the variation in small signal equivalent circuit model parameters for a Si/SiGe HBT process

On a single wafer and between different wafers, there are variations in device characteristics due to processing non-uniformity and non-reproducibility. In order to capture these variations, we could extract a complete equivalent circuit model for each device. Since this is a timeconsuming process, we developed an alternative approach based on an Artificial Neural Network (ANN). This ANN takes measured quantities as inputs, and generates the model parameters as outputs. To keep the complexity of the ANN down to a reasonable size, we limited this mapping to the most sensitive elements. These are determined by performing a sensitivity analysis on a reference device. To demonstrate the method, we applied it to a Si/SiCe HBT process. Results show that the ANN very well predicts model parameters and that it is B very good statistical model. Index Terms De-embedding, small-signal modeling, artificial neural networks, Statistical Database, SiCe HBT.

[1]  J. Purviance,et al.  Statistical design for microwave systems , 1991, 1991 IEEE MTT-S International Microwave Symposium Digest.

[2]  C. Potratz,et al.  An efficient linear statistical FET model , 1989, IEEE MTT-S International Microwave Symposium Digest.

[3]  J. Carroll,et al.  FET parameter orthogonalization with principal components , 1994, 1994 IEEE MTT-S International Microwave Symposium Digest (Cat. No.94CH3389-4).

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

[5]  P. J. Tasker,et al.  A physical, yet simple, small-signal equivalent circuit for the heterojunction bipolar transistor , 1997 .

[6]  M.C.E. Yagoub,et al.  A robust algorithm for automatic development of neural network models for microwave applications , 2001, 2001 IEEE MTT-S International Microwave Sympsoium Digest (Cat. No.01CH37157).

[7]  E. Strid,et al.  GHz on-silicon-wafer probing calibration methods , 1988, Proceedings of the 1988 Bipolar Circuits and Technology Meeting,.

[8]  M. Fernandez-Barciela,et al.  HBT small signal T and /spl pi/ model extraction using a simple, robust and fully analytical procedure , 2002, 2002 IEEE MTT-S International Microwave Symposium Digest (Cat. No.02CH37278).

[9]  Qi-Jun Zhang,et al.  Neural Networks for RF and Microwave Design , 2000 .

[10]  Fadhel M. Ghannouchi,et al.  A new analytical and broadband method for determining the HBT small-signal model parameters , 2000, 2000 IEEE MTT-S International Microwave Symposium Digest (Cat. No.00CH37017).

[11]  W. Heinrich,et al.  Nonlinear modelling of SiGe HBTs up to 50 GHz , 1997, 1997 IEEE MTT-S International Microwave Symposium Digest.

[12]  F Vallette,et al.  Some results obtained in sensitivity analysis of microwave circuits , 1996, 1996 26th European Microwave Conference.