[SI-LIST] Re: rational function approximation
- From: Amitava Bhaduri <bhadhua@xxxxxxxxxxxx>
- To: Ray Anderson <Raymond.Anderson@xxxxxxx>
- Date: Thu, 15 Apr 2004 23:52:24 -0400 (EDT)
Hi Ray,
I haven't extrapolated the S-parameters down to DC(s=0). I have generated
the Z-parameters(and thereby S-parameters) of the the interconnects
starting at 1GHz to 4GHz at uniform spacing (10 samples). I was trying to
do the eigenvalue drill but as I do not have enough experience in this I
am not sure how to deal with 10 different values of each of Z11, Z12, Z21
and Z22(for 2-port interconnect structure) and combine them to a rational
function of the form P(s)/Q(s). After I get to this point I have to find a
rational function approximation that takes care of the passivity problem
and reduces the model. I can then possibly find an equivalent circuit that
can be plugged into the netlist and may be able to avoid using "nport". I
just gathered the basic idea by reading some papers, but don't know the
drill to make this work. Any help would be great.
I guess "nport" does interpolation (by specifying the option
interp=rational), but probably fails to take care of the
passivity/stability of the rational function.
Thanks,
Amitava
On Thu, 15 Apr 2004, Ray Anderson wrote:
> Amitava-
>
> Question: Did you extrapolate your s-parameter data down to DC (or
> otherwise define the s-parameters at zero frequency)? I have been told
> that when creating rational function approximations from s-parameters
> that the s-parameter behavior at DC can effect the passivity of the
> resultant model. Did you go through the eigenvalue drill to prove or
> disprove passivity or does Spectre check it for you?
>
> Does Spectre actually do a rational function approximation internally
> with Laplace type elements or is it creating a model using a frequency
> response table ?
>
> I'm not familiar with the nport element in Spectre. Do you know if it
> is it similar to the Hspice s-element ?
>
>
> Maybe some of the modeling gurus from the CAD tool companies will
> respond with a better answer.
>
> -Ray Anderson
> Sun Microsystems Inc.
>
>
> Amitava Bhaduri wrote:
> >
> > Hi,
> >
> > Using a full-wave tool I have been able to generate an impedance matrix of
> > an N-port interconnect and later converted the Z matrix into S-parameters.
> > Using "nport" from spectre I tried connecting the generated parametric
> > data to the remaining device netlist of the circuit. But the simulation
> > doesn't work due to stability and passivity problem. I have generated the
> > impedance data for 10 frquency points.
> > I have learnt that using rational function approximation the passivity of
> > the network can be ensured, but I am not sure how to proceed with that
> > with Z-matrix (or S-matrix) in hand. Can anyone suggest ways or the
> > necessary steps that I have to folow in order to make the interconnect
> > network passive and use it with spectre circuit simulator. Is it possible
> > to use MATLAB to do this rational function approximation and if it is,
> > then how do I do it?
> >
> > Thanks in advance,
> > Amitava
> >
> > Amitava Bhaduri
> > Research Assistant
> > 527 ERC
> > University of Cincinnati
> > OH - 45220
> > ph(lab) : 513-556-3025
>
>
> ------------------------------------------------------------------
> To unsubscribe from si-list:
> si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
>
> or to administer your membership from a web page, go to:
> //www.freelists.org/webpage/si-list
>
> For help:
> si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
>
> List FAQ wiki page is located at:
> http://si-list.org/wiki/wiki.pl?Si-List_FAQ
>
> List technical documents are available at:
> http://www.si-list.org
>
> List archives are viewable at:
> //www.freelists.org/archives/si-list
> or at our remote archives:
> http://groups.yahoo.com/group/si-list/messages
> Old (prior to June 6, 2001) list archives are viewable at:
> http://www.qsl.net/wb6tpu
>
>
------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
or to administer your membership from a web page, go to:
//www.freelists.org/webpage/si-list
For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
List FAQ wiki page is located at:
http://si-list.org/wiki/wiki.pl?Si-List_FAQ
List technical documents are available at:
http://www.si-list.org
List archives are viewable at:
//www.freelists.org/archives/si-list
or at our remote archives:
http://groups.yahoo.com/group/si-list/messages
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
Other related posts:
