A yield centric statistical design method for optimization of the SRAM active column

For robust design of SRAM memories, it is not sufficient to guarantee good statistical margins on the SRAM cell parameters. The sense amplifier needs sufficient input signal before it can reliably sense the data, while the SRAM cell requires sufficient time to develop that input signal. This paper p...

Full description

Saved in:
Bibliographic Details
Main Authors: Doorn, T.S., Croon, J.A., ter Maten, E.J.W., Di Bucchianico, A.
Format: Conference Proceeding
Language:English
Subjects:
Design methodology
Design optimization
Optimization methods
Probability
Random access memory
Robustness
Statistical analysis
Stochastic processes
System performance
Transistors
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:For robust design of SRAM memories, it is not sufficient to guarantee good statistical margins on the SRAM cell parameters. The sense amplifier needs sufficient input signal before it can reliably sense the data, while the SRAM cell requires sufficient time to develop that input signal. This paper presents a new statistical method that allows optimization of the access time of an SRAM memory, while guaranteeing a yield target set by the designer. Using this method, the access time of a high performance advanced CMOS SRAM has been improved 6%, while simultaneously reducing the sense amplifier size.
ISSN:1930-8833
2643-1319
DOI:10.1109/ESSCIRC.2009.5325954