Statistical circuit characterization for deep-submicron CMOS designs

James Chen*, Michael Orshansky, Chen-Ming Hu, C. P. Wan

*Corresponding author for this work

Research output: Contribution to journalConference articlepeer-review

6 Scopus citations


Aggressive scaling of CMOS MOSFET gate lengths is halving minimum device channel lengths every 4 to 6 years. This brings an increase in deep submicron device performance sensitivity to manufacturing variations. Its impact on highly-integrated, system-on-a-chip designs needs assessment to ensure first-pass silicon and adequate yield. Standard simulations to assess these impacts are performed at SPICE simulation level and rely upon a multivariate statistical technique called principal component analysis. This method is popular because it reduces the large number of correlated SPICE model parameters characteristic of current deep-submicron models into few statistically independent factors which can then be generated by Monte Carlo simulation.

Original languageEnglish
Pages (from-to)90-91
Number of pages2
JournalDigest of Technical Papers - IEEE International Solid-State Circuits Conference
StatePublished - 1 Jan 1998
EventProceedings of the 1998 IEEE 45th International Solid-State Circuits Conference, ISSCC - San Francisco, CA, USA
Duration: 5 Feb 19987 Feb 1998

Fingerprint Dive into the research topics of 'Statistical circuit characterization for deep-submicron CMOS designs'. Together they form a unique fingerprint.

Cite this