A systematic and physical application of multivariate statistics to MOSFET I-V models

Author

Kondo, Masaki ; Onodera, Hidetoshi ; Tamaru, Keikichi

Author_Institution

Micro-Electron. Dev. Center, Toshiba Corp., Yokohama, Japan

fYear

1999

fDate

1999

Firstpage

34

Lastpage

37

Abstract

A statistical method applicable to MOSFET compact models for circuit simulation is proposed. A salient feature of the method is that correlation between device parameters is formulated by independent physical parameters which dominate MOSFET characteristics. The key idea is the introduction of an intermediate model. With the use of the intermediate model, physical parameter fluctuations are systematically mapped into the parameters of many device models. As the device parameters are expressed as functions of the independent physical parameters, the worst-case parameters can be accurately derived from the statistical model. The efficiency of the proposed method is shown with experimental results from a 0.3 μm CMOS processing technology

Keywords

CMOS integrated circuits; MOSFET; circuit simulation; electric current; semiconductor device models; statistical analysis; 0.3 micron; CMOS processing technology; MOSFET I-V models; MOSFET characteristics; MOSFET compact models; circuit simulation; device model parameters; device models; device parameter correlation; independent physical parameters; intermediate model; multivariate statistics; physical multivariate statistics application; physical parameter fluctuation systematic mapping; physical parameter fluctuations; statistical method; statistical model; systematic multivariate statistics application; worst-case parameters; CMOS process; CMOS technology; Circuit simulation; Fluctuations; Independent component analysis; Integrated circuit modeling; MOSFET circuits; Semiconductor device modeling; Statistical analysis; Statistics;

fLanguage

English

Publisher

ieee

Conference_Titel

Statistical Metrology, 1999. IWSM. 1999 4th International Workshop on

Conference_Location

Kyoto

Print_ISBN

0-7803-5154-1

Type

conf

DOI

10.1109/IWSTM.1999.773190

Filename

773190