A novel approach to statistical modeling using cumulative probability distribution fitting

Author

Bandler, J.W. ; Biernacki, R.M. ; Cai, Q. ; Chen, S.H.

Author_Institution

Dept. of Electr. & Comput. Eng., McMaster Univ., Hamilton, Ont., Canada

fYear

1994

fDate

23-27 May 1994

Firstpage

385

Abstract

A novel approach to statistical modeling is presented. The statistical model is directly extracted by fitting the cumulative probability distributions (CPDs) of the model responses to those of the measured data. This new technique is based on a solid mathematical foundation and, therefore, should prove more reliable and robust than the existing methods. The approach is illustrated by statistical MESFET modeling based on a physics-oriented model which combines the modified Khatibzadeh and Trew model and the Ladbrooke model (KTL). The approach is compared with the established parameter extraction/postprocessing approach (PEP) in the context of yield verification.<>

Keywords

Schottky gate field effect transistors; probability; semiconductor device models; solid-state microwave devices; statistical analysis; Ladbrooke model; MESFET modeling; cumulative probability distribution fitting; model responses; modified Khatibzadeh/Trew model; physics-oriented model; statistical modeling; yield verification; Data mining; MESFETs; Parameter extraction; Probability distribution; Robustness; Solids;

fLanguage

English

Publisher

ieee

Conference_Titel

Microwave Symposium Digest, 1994., IEEE MTT-S International

Conference_Location

San Diego, CA, USA

ISSN

0149-645X

Print_ISBN

0-7803-1778-5

Type

conf

DOI

10.1109/MWSYM.1994.335413

Filename

335413