The best way to deal with the time offset may be to take the data between the
20% and the 80% points of the rising edge (if there is one) of each waveform
and use standard curve fitting techniques to develop two "best fit" equations.
Then time adjust one set of data for a minimum delta in time between the two
equations. And maybe that delta is also determined by subtraction,
only t2-t1 instead of v2-v1. This technique could also be useful to compare
two sets of measured data from parts that are supposed to be from the same lot.
But as Tom Dagostino said earlier, any random measured part should only be
checked to make sure it is more than the minimum and less than the max.
Dima Smolyansky wrote:
> One of the reasons that the human eye is such a great processing device is
> that it selects which portions of the waveform are important and which are
> not. Emulating this process, perhaps subtraction using a weight function is
> a better way that direct subtraction. Then again, as Alex pointed out, one
> has to deal with any time offset between the measured and the simulated
> Dima Smolyansky
> Time Domain Analysis Systems, Inc.
> 7465 SW Elmwood St.
> Portland, OR 97223
> (503) 977-3629
> (503) 804-7171 (mobile)
> (503) 245-5684 (fax)
> The Interconnect Modeling Company (TM)
> -----Original Message-----
> From: firstname.lastname@example.org <email@example.com>
> To: firstname.lastname@example.org <email@example.com>
> Date: Thursday, July 29, 1999 3:14 PM
> Subject: Re: [SI-LIST] : Waveform comparison metrics
> >Funny you should post this today - we were just working on this very thing!
> >application is a method to compare lab data with simulation predictions
> >IBIS or SPICE models, whichever you happen to have. I am part of a
> >that wrote a document called the "IBIS Accuracy Specification," which you
> >find on the IBIS web site under accuracy, if you're interested. We settled
> on a
> >twist on your first method, subtraction. First you have to interpolate to
> >the waveforms on a common x-axis. (Scopes don't usually allow you to pick
> >while most simulators do.) Then you have to slide the waveforms so that
> >both cross some threshold at the same time. Finally you take the absolute
> >of the difference between two data points and average these numbers over
> >whole set of data points. The method does have its down side. A dc offset
> >make the correlation look worse than it really is. Likewise, really good
> >correlation on one part of the waveform (say, the steady state) can mask
> >lousy correlation on another part of the waveform.
> >I'd be interested in hearing more about your correlation coefficient idea.
> >best solution may be to run 2 or 3 metrics, along with a listing of what
> >call "basic metrics." Let each metric tell you something different about
> >correlation. It would be real nice if we had a piece of shareware to do
> >computations. We're leaning toward having our IC vendors provide us with
> >kind of data so we can assess model quality. It would be good if everyone
> >singing from the same hymnal!
> >Anybody else have other ideas?
> >Greg Edlund
> >Advisory Engineer, Critical Net Analysis
> >3650 Hwy. 52 N, Dept. HDC
> >Rochester, MN 55901
> >"Levin, Alexander" <firstname.lastname@example.org> on 07/29/99 02:06:22 PM
> >Please respond to email@example.com
> >To: "'firstname.lastname@example.org'" <email@example.com>
> >Subject: [SI-LIST] : Waveform comparison metrics
> >In the course of a design, there are many occasions requiring the
> >of waveforms to determine "sameness". Several tasks requiring waveform
> >comparison come to mind: building and checking IBIS models,
> >comparing/benchmarking simulator tools, and comparing simulated waveforms
> >lab measurements. Aside from the basic metrics of rising/falling delay,
> >ringback amplitude, overshoot/undershoot, is there a repeatable
> >(automatable?) method which can quantify the degree of matchup between two
> >Several approaches come to mind, but each carries its own downfalls.
> >Waveform subraction: Will provide an estimate of the differences between
> >waveforms, but is extremely sensitive to any time offset.
> >Correlation coefficient: Analagous to a dot product of the voltage-time
> >sample points in each waveform. Doesn't capture absolute DC voltage level
> >shifts; r^2 offers no information about type of mismatch.
> >FFT: Can capture similarity in edge rates, ringing period, etc. but key
> >waveform differences may be masked or lost in high frequency noise seen in
> >FFT's of digital waveforms.
> >Overlaying and eyeballing: The human eye is an excellent image processing
> >device, but the statements "looks good" or "that's a lot of overshoot" are
> >often too subjective.
> >Apply basic metrics: Measuring rise/fall delay, ringback, over/undershoot,
> >ringing period, etc. are typically used. Again, however, the
> >of the matchup is subjective.
> >I'm not necessarily looking for a catchall solution, but would be
> >interesting in hearing about any novel approaches people are using.
> >Waveform overlay will still have its place, but it would be nice to combine
> >this with less subjective methods.
> >Thanks much,
> >Alex Levin
> >**** To unsubscribe from si-list: send e-mail to
> firstname.lastname@example.org. In
> >the BODY of message put: UNSUBSCRIBE si-list, for more help, put HELP.
> >archives are accessible at http://www.qsl.net/wb6tpu/si-list ****
> >**** To unsubscribe from si-list: send e-mail to
> email@example.com. In the BODY of message put: UNSUBSCRIBE
> si-list, for more help, put HELP. si-list archives are accessible at
> http://www.qsl.net/wb6tpu/si-list ****
> **** To unsubscribe from si-list: send e-mail to firstname.lastname@example.org. In the BODY of message put: UNSUBSCRIBE si-list, for more help, put HELP. si-list archives are accessible at http://www.qsl.net/wb6tpu/si-list ****
**** To unsubscribe from si-list: send e-mail to email@example.com. In the BODY of message put: UNSUBSCRIBE si-list, for more help, put HELP. si-list archives are accessible at http://www.qsl.net/wb6tpu/si-list ****