Title :
Statistical modeling tools, methods and applications for integrated circuit manufacturability
Author :
Iravani, Farshid ; Habu, Masa ; Khalily, Ebrahim
Author_Institution :
Hewlett-Packard Co., Santa Clara, CA, USA
Abstract :
A description of various statistical modeling methods is provided. The analysis methods are discussed to remove the confusion in the existing literature. Statistical models are generated using factor analytic technique for SPICE level 3 on a 0.8 micron LDD CMOS. It is shown that our measurement based approach accurately predicts the device performance. We show that the existence of a “physical MOS model” is not a necessary pre-requisite to perform statistical modeling. Our approach provides models that are suitable for both analog and digital designs and promises to make statistical modeling feasible for general use
Keywords :
CMOS analogue integrated circuits; CMOS digital integrated circuits; SPICE; integrated circuit design; integrated circuit modelling; semiconductor process modelling; statistical analysis; 0.8 micron; LDD CMOS; SPICE level 3; analog design; device performance; digital design; factor analytic technique; integrated circuit manufacturability; physical MOS model; statistical modeling tools; Application specific integrated circuits; Circuit optimization; Data analysis; Fluctuations; Integrated circuit manufacture; Integrated circuit modeling; Principal component analysis; Semiconductor device modeling; Statistical analysis; Virtual manufacturing;
Conference_Titel :
Microelectronic Test Structures, 1995. ICMTS 1995. Proceedings of the 1995 International Conference on
Conference_Location :
Nara
Print_ISBN :
0-7803-2065-4
DOI :
10.1109/ICMTS.1995.513973