The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> Analog Design >> Correlation coefficient in Montecarlo simulation
https://designers-guide.org/forum/YaBB.pl?num=1118309223

Message started by fran2k5 on Jun 9^th, 2005, 2:27am

Title: Correlation coefficient in Montecarlo simulation
Post by fran2k5 on Jun 9^th, 2005, 2:27am

Hello everybody,

I am performing some Montecarlo simulations in Cadence and I would like to know how can I estimate correlation coefficient between matched devices.

Some devices are going to be laid out very close trying to match them as much as possible. As a consequence, I think that considering them completely uncorrelated in the Montecarlo sims is going to give too pessimistic result. It is clear that the correlation coefficient depends on the layout but it would be nice to have educated guesses of the correlation coefficient vs distance. Do manifacturers have this kind of infos?

I would appreciate any comment on this.

Thank you very much,

frank

Title: Re: Correlation coefficient in Montecarlo simulati
Post by Geoffrey_Coram on Jun 23^rd, 2005, 8:07am

The foundries may have this information, but whether they will let you have it may be another question. Also, the correlation may depend very strongly on the layout -- how closely did you match the poly environment (with dummy fingers, etc.)?

Title: Re: Correlation coefficient in Montecarlo simulati
Post by fran2k5 on Jun 23^rd, 2005, 11:36am

Thank you very much for your answer.

It would be nice to know at least the most common values of correlation coefficient for typical conditions.
I would like to have an educated guess for the correlation coefficient. As I said there a lot of situations in which devices can not be considered completely uncorrelated (differential pairs, etc) just because they are laid out really close.

We can assume that special layout techniques (common centroid, ect) are going to improve the matching between devices.