Title :
Parameter Estimate by Incomplete Neural Network for MOSFET Life Model with NOME Distribution
Author :
Tong, Qiaoling ; Zou, Xuecheng ; Tong, Hengqing ; Liu, Tianzhen
Author_Institution :
Huazhong Univ. of Sci. & Technol., Wuhan
Abstract :
The test of MOSFET life is an important work in semiconductor manufacture. Many studies have been done on modeling the MOSFET life distribution for the reliability of integrated circuits. In this paper, a MOSFET life model is introduced by making use of threshold voltage drift property. A new statistical distribution, whose negative order moment estimator is just the maximum likelihood estimator, is deduced, and we call it NOME distribution. A new method of parameter estimate, F estimate of parameter, is proposed for complete sample test and truncated sample test. A kind of incomplete neural network is used to optimize the parameter in the model. Results show that neural network is a powerful tool for the analysis of MOSFET life model.
Keywords :
MOSFET; electronic engineering computing; integrated circuit reliability; maximum likelihood estimation; neural nets; statistical distributions; MOSFET life model; NOME distribution; integrated circuit reliability; maximum likelihood estimator; negative order moment estimator; neural network; parameter estimate; semiconductor manufacture; statistical distribution; threshold voltage drift property; Circuit testing; Integrated circuit modeling; Integrated circuit reliability; Life estimation; Life testing; MOSFET circuits; Neural networks; Parameter estimation; Semiconductor device manufacture; Semiconductor device testing; F; Incomplete neural network.; MOSFET life model; NOME distribution; parameter estimation;
Conference_Titel :
Natural Computation, 2007. ICNC 2007. Third International Conference on
Conference_Location :
Haikou
Print_ISBN :
978-0-7695-2875-5
DOI :
10.1109/ICNC.2007.538
