[SI-LIST] Re: rms jitter vs p-p jitter
- From: "Ashok Babu K" <k.ashokbabu@xxxxxxxxxxxxx>
- To: <mpeng@xxxxxxxxxxxxx>, <alexh@xxxxxxxxxxxxxxxx>,<hata@xxxxxxxxxx>, <si-list@xxxxxxxxxxxxx>
- Date: Thu, 4 Oct 2001 16:35:12 +0530
Hi All,
A simple conversion factor will be p-p jitter = rms jitter * x. As every
one has explained the value of 'x' is to be derived from the probability
level. If the probability distribution of jitter is considered as Gaussian,
there is no gurantee to specify the value of p-p jitter that bounds the
jitter always. The reason is the Gaussian distribution extends indefinitely
on both the sides of the mean value. In short, as Mike has said, we can be
sure that the value of the p-p jitter we get from rms jitter value bounds
99.9% of the time, if we take the value of 'x' as 6.
So the idea is to quantify the probability level to something that is more
of our concern. As Brad has indicated from the FC standard, one obvious
parameter will be to attach the probality level is BER.
The Maxim application note is really a good source for this. This app note
can be found in Maxim website as 3hfan402.pdf.
Regards,
Ashok.
----- Original Message -----
From: "Mike Li" <mpeng@xxxxxxxxxxxxx>
To: <alexh@xxxxxxxxxxxxxxxx>; <hata@xxxxxxxxxx>; <si-list@xxxxxxxxxxxxx>
Cc: "Mike Li" <mpeng@xxxxxxxxxxxxx>
Sent: Saturday, September 29, 2001 1:36 AM
Subject: [SI-LIST] Re: rms jitter vs p-p jitter
>
> Appreciate if somebody can shed some light:
> If given a rms jitter value, is there any way this can be
> converted into
> peak-peak jitter value?
>
> The answer is NO in general!
>
> peak-to-peak jitter is a wrong term to be used for jitter measurement
> due to the presence of unbounded random jitter. Random jitter, in general,
> has a Gaussian distribution. Jitter has two components, one is
> deterministic jitter (DJ) and is bounded, another is the random jitter
(RJ)
> and is unbounded. peak-to-peak should be used to only for DJ, while
> rms for RJ.
>
> From a thoretical point view, if the jitter distribution is a Gaussian,
> ther is a way to relate the rms to peak-to-peak jitter only if the
> probabalibity
> level is given. For example. At a probability level of ~ 1e-3,
peak-to-peak
> is ~ 6sigma; at a probabality level of ~1e-12, peak-to-peak is ~ 14 sigma.
>
> Wavecrest has done considerable works in jitter analysis and measurement
in
> the past
> a few years. The web site is http://www.wavecrestcorp.com/
>
> Also, I have attached a review paper on jitter that was presented in
> designcon99
>
>
> Hope this helps.
>
> Mike
>
>
> > -----Original Message-----
> > From: Alex Horvath [mailto:alexh@xxxxxxxxxxxxxxxx]
> > Sent: Friday, September 28, 2001 11:56 AM
> > To: alexh@xxxxxxxxxxxxxxxx; hata@xxxxxxxxxx; si-list@xxxxxxxxxxxxx
> > Subject: [SI-LIST] Re: rms jitter vs p-p jitter
> >
> >
> >
> > Ok, RMS is essentially the same as sigma. Three sigma
> > includes 99.8% of all
> > samples. So you would probably be safe in saying that the
> > peak jitter is 3
> > times the RMS.
> >
> > -----Original Message-----
> > From: si-list-bounce@xxxxxxxxxxxxx
> > [mailto:si-list-bounce@xxxxxxxxxxxxx]On Behalf Of Alex Horvath
> > Sent: Friday, September 28, 2001 11:44 AM
> > To: hata@xxxxxxxxxx; si-list@xxxxxxxxxxxxx
> > Subject: [SI-LIST] Re: rms jitter vs p-p jitter
> >
> >
> >
> > For a non-random process peak 1/sqrt2 * VRMS however jitter
> > is a random
> > process so in theoretical terms the peak jitter value has no
> > meaning. You
> > have to define it in terms of sigma (assuming it's a normal
> > dist). There
> > must be some math gurus here that could relate RMS to sigma.
> > Then you can
> > say something like "the 3 sigma jitter is x ns", 3 sigma
> > including something
> > like 98% (I forgot the exact figure) of all jitter samples.
> >
> > -----Original Message-----
> > From: si-list-bounce@xxxxxxxxxxxxx
> > [mailto:si-list-bounce@xxxxxxxxxxxxx]On Behalf Of hata@xxxxxxxxxx
> > Sent: Friday, September 28, 2001 4:58 AM
> > To: si-list@xxxxxxxxxxxxx
> > Subject: [SI-LIST] rms jitter vs p-p jitter
> >
> >
> >
> > Dear SI guru,
> >
> > Appreciate if somebody can shed some light:
> > If given a rms jitter value, is there any way this can be
> > converted into
> > peak-peak jitter value?
> >
> > Thank you.
> > _____________________________________________________________
> >
> >
> > ------------------------------------------------------------------
> > 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 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 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 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 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 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:
