DocumentCode
1996608
Title
An analytical mismatch model of nano-CMOS device under impact of intrinsic device variability
Author
Hong, Feng ; Cheng, Binjie ; Roy, Scott ; Cumming, David
Author_Institution
Dept. of Electron. & Electr. Eng., Univ. of Glasgow, Glasgow, UK
fYear
2011
fDate
15-18 May 2011
Firstpage
2257
Lastpage
2260
Abstract
We present for the first time an analytical model for the effect of short-gate transistor mismatch on analogue circuit design. Analogue circuit design is very vulnerable to device mismatch, as large numbers of matching-sensitive circuits are used. This is particularly severe in short-gate CMOS processes. In this paper, a new analytical mismatch model is developed for both triode and saturation regimes and verified using 35nm gate-length BSIM4 model cards. Short-channel effects, such as velocity saturation and mobility degradation, are taken into consideration. The results achieved excellent agreement with Monte Carlo HSPICE simulations. Furthermore, this model can be used to develop a rapid estimate of the precision or production yield of a circuit based only on the process statistics.
Keywords
CMOS analogue integrated circuits; MOSFET; Monte Carlo methods; SPICE; integrated circuit design; nanoelectronics; semiconductor device models; statistics; triodes; Monte Carlo HSPICE simulation; analogue circuit design; analytical mismatch model; gate-length BSIM4 model cards; intrinsic device variability; matching-sensitive circuits; nanoCMOS device; process statistics; saturation regimes; short-channel effects; short-gate CMOS processes; short-gate transistor effect; triode; Analytical models; Degradation; Integrated circuit modeling; Logic gates; Mathematical model; Monte Carlo methods; Transistors;
fLanguage
English
Publisher
ieee
Conference_Titel
Circuits and Systems (ISCAS), 2011 IEEE International Symposium on
Conference_Location
Rio de Janeiro
ISSN
0271-4302
Print_ISBN
978-1-4244-9473-6
Electronic_ISBN
0271-4302
Type
conf
DOI
10.1109/ISCAS.2011.5938051
Filename
5938051
Link To Document