Stats

  • Locked Locked
  • Replies 5
  • Subscribers 125
  • Views 17640
  • Members are here 0
This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

linear fit of simulation results

olalivier
In ADE I extract some parameters from a waveform (time elapsed from rising to falling edge on some kind of TOT) on a parametric simulation. I plot the result but I would like to get a linear fit of this curve plus residuals. I can't find a way to do that with in the calc. Can you please tell me if that fonction exists. Thank you very much Olivier
  • Andrew Beckett

    There's nothing built-in currently, although it wouldn't be too hard to write a function that performed linear regression on the points in the curve and produced a best-fit line using a least-squares method. You could then subtract this from the original curve to get the residuals. I'd write it myself but don't really have time right now to knock something together.

    There is an existing enhancement CCR, 57477, which is for doing a polynomial fit in the calculator, but that's not been implemented yet (it's a long-standing low-priority enhancement request with only 2 requests over 13 years, so not that likely to get implemented any time soon).

    Regards,

    Andrew.

  • olalivier
    I don't want to sound sarcastic but you mean we're only 3 in the world doing fits on result...!!?? I really feel that it should be a built-in function but obviously I'm wrong. Though this kind of algorithm exists in many math libs like gsl, for example.

    For now I guess I'll stick to external tools, which is not very convenient to get this kind of result...
  • Andrew Beckett

    I suspect most people doing this have just done it externally rather than wanting to repeatedly do it in ADE. To me it seems an entirely reasonable request, but we have to prioritise what people actually ask for, of course... otherwise it's so easy to develop features that we think people want and then find nobody in practice is using it.

    I'll put this on my list of things to put together when I have a spare moment (unfortunately it's quite a long list!)

    Regards,

    Andrew.

  • JEFO
    in reply to Andrew Beckett

    Hello Andrew

    Any news on this particular feature?

    I've been missing such a feature as a linear fit to datapoints, meant for extracting muCox and threshold voltages. 

    But as you suggest, it's not entirely too hard to write one yourself.


    Best regards

    Jacob

  • Andrew Beckett
    in reply to JEFO

    Hi Jacob,

    Strangely enough I wrote some code to do this less than two weeks ago - as I said in this other post (where you can find the code), I knew it had come up in the past (I'd lost track of where though...)

    I also published it as an article on Cadence Online Support and it's been added to the Custom IC Calculator SKILL Function Library.

    Regards,

    Andrew.

Community Guidelines

The Cadence Design Communities support Cadence users and technologists interacting to exchange ideas, news, technical information, and best practices to solve problems and get the most from Cadence technology. The community is open to everyone, and to provide the most value, we require participants to follow our Community Guidelines that facilitate a quality exchange of ideas and information. By accessing, contributing, using or downloading any materials from the site, you agree to be bound by the full Community Guidelines.