Including geometric variation of transistors using Montecarlo simulation in Cadence

[Junus2012]

Hello friends,

I read couple of papers in which the author performed a montecarlo based on geometric (dimention) variations. How can I define such setup ? is there any setting for that ?

for further detail you can see the following papers

https://ieeexplore.ieee.org/document/4349224?arnumber=4349224

https://link.springer.com/article/10.1007/s10470-016-0874-2


Also in the first paper the author run the Montecarlo in conjuction of corners (SS, FF...etc), why he should include these corners when it will aready covered by the montecarlo variation ?

Thank you

 

[pancho_hideboo]

Simply use mismatch parameters which include geometrical dependencies.

See https://designers-guide.org/forum/YaBB.pl?num=1276330403

 

[ThisIsNotSam]

Also in the first paper the author run the Montecarlo in conjuction of corners (SS, FF...etc), why he should include these corners when it will aready covered by the montecarlo variation ?

Do you have any model you can play with? The answer to this question becomes quite trivial when you see how foundries organize their models, corners, and process variation distinction (global and local)

 

[Dominik Przyborowski]

Simply, read this friendly manual. In this case, your process documentation. For your process is nice and short document about Mismatch and how it is modeled as well.