RE: [802.3ae] XAUI Rj TR comment
Hello Pat,
It's good that we are having this conversation about crosstalk jitter. This
is a subject that comes up often in our MJSQ meetings. The model we have
used for crosstalk has been one of simple mutual indunctance and capacitive
coupling, two traces on a PCB very close to one another for example. It
sounds like you have a better model that you are using?
Also, We used the PRBS patterns vs. an 8b/10b pattern to try and create less
deterministic interference and if we where to make that pattern long enough,
2e31-1, for example would probably create some very low probablity DJ, as
you mentioned, truncated gaussian. In any event that experiment has yet to
be performed showing crosstalk that creates any significant ammount of
gaussian jitter. Pat, speaking for the T11, MJSQ group, we would welcome
your input in this matter.
Regrads,
Dennis
-----Original Message-----
From: THALER,PAT (A-Roseville,ex1)
To: Dennis Petrich; THALER,PAT (A-Roseville,ex1); Lindsay, Tom; Howard A.
Baumer; HSSG_reflector (E-mail)
Sent: 8/8/2001 4:48 PM
Subject: RE: [802.3ae] XAUI Rj TR comment
Dennis,
I have now looked at the crosstalk documents T11/00-064v0 and
T11/99-759v0.
They create "crosstalk" by taking a PRBS signal running at near the baud
rate of the data signal and attenuating before adding it to the data
signal.
This would be a good model of crosstalk only if crosstalk coupling had
uniform phase and gain vs. frequency. In that case the crosstalk from a
disturbing signal would look just like an attenuated version of the
attenuated signal. Normally crosstalk coupling is nothing like an
attenuator. Crosstalk gain from common crosstalk mechanisms stays under
an
envelope that rises with frequency but the gain bounces around under
that
envelope in a manner that looks random. The phase also varies wildly
with
frequency.
So, the crosstalk amplitude from a disturber is the sum of all the
frequency
components from the disturber coupling with varying frequency and phase.
At
some times, those components are canceling each other and at other times
they are adding. The crosstalk signal looks like "noise" with amplitude
that
varies over time and not like a PRBS signal with transitions that vary
but a
relatively uniform amplitude.
Of course if one couples an attenuated PRBS signal into the data signal,
one
creates Bounded Uncorrelated Jitter. The disturbing signal in the tests
is
contributing noise that is set somewhere between 10% and 20% of the data
amplitude. It can only move an zero crossing while the signal edge is
within
that distance of the zero crossing.
That does nothing to show that crosstalk provides jitter that is bounded
(where unbounded means the data is Gaussian or truncated Gaussian where
the
truncation happens past 1E-12).
Regards,
Pat
-----Original Message-----
From: Dennis Petrich [mailto:dpetrich@xxxxxxxxxxxxx]
Sent: Wednesday, August 01, 2001 2:11 PM
To: 'THALER,PAT (A-Roseville,ex1)'; Lindsay, Tom; Howard A. Baumer;
HSSG_reflector (E-mail)
Subject: RE: [802.3ae] XAUI Rj TR comment
Pat,
The NCITS "TR-25-1999" MJS document provides definitions that can be
referenced for XAUI use to distinguish between the various jitter types
such
as RJ, DJ, DDJ, SJ and so on.
Also, FC crosstalk work was done a while back and can be viewed at
T11/00-064v0 and T11/99-759v0. In the tests crosstalk showed up as DJ.
But
I'm sure these results would vary as a function of the crosstalking data
rate and frequency content.
Dennis
-----Original Message-----
From: THALER,PAT (A-Roseville,ex1) [mailto:pat_thaler@xxxxxxxxxxx]
Sent: Wednesday, August 01, 2001 1:43 PM
To: Lindsay, Tom; Howard A. Baumer; HSSG_reflector (E-mail)
Subject: RE: [802.3ae] XAUI Rj TR comment
Tom,
Jitter always seems to be a difficult subject to sort out and your
remark
below caused me to do some checking on RJ vs. DJ.
I've looked all through the 802.3 standard and our draft. There doesn't
seem
to be any definition of RJ or DJ. Processes can certainly be random
without
being random or Gaussian. Deterministic means if a set of conditions is
set
up we know what will result. The roll of a die is random though the
result
is bounded.
If we are using dictionary words with a different or more restricted
meaning
such as random = Gaussian (or truncated Gaussian where the truncation
happens past 1E-12) then we should define our terms. Since we specify
deterministic jitter and total jitter, we should at least have a
reasonably
rigorous definition of "deterministic jitter."
I also notice that in some places jitter is divided into RJ and DJ, but
in
other places in 47 it is RJ, DJ and sinusoidal. 52.9.10.4 (and the
equivalent subclause of 53) divide jitter into random, deterministic and
bounded.
Crosstalk is deterministic in that given a fixed adjacent signal and a
fixed
coupling function one can determine the crosstalk. However, the
crosstalk at
a receiver is often the result of multiple disturbers coupling in each
with
its own function and the signals aren't correlated to the received
signal.
Therefore, the sum of the crosstalk looks like a truncated Gaussian.
Even if
the definition of RJ is Gaussian up to at least 1E-12, it isn't clear to
me
that crosstalk would fall outside that definition. I don't recall seeing
any
studies on the distribution of crosstalk for XAUI or for our optical
receivers.
I would expect crosstalk to be part of RJ rather than DJ.
Regards,
Pat
-----Original Message-----
From: Lindsay, Tom [mailto:tlindsay@xxxxxxxxxxxxxxxxxxxx]
See below, Tom
-----Original Message-----
From: Howard A. Baumer [mailto:hbaumer@xxxxxxxxxxxx]
Sent: Tuesday, July 31, 2001 11:36 AM
To: Lindsay, Tom; HSSG_reflector (E-mail)
Subject: Re: [802.3ae] XAUI Rj TR comment
>>>>snip<<<<<
- We're still confused on how you would ever get 0.55UI of RJ. If
crosstalk adds so much jitter,
**TL - crosstalk is expected to be bounded, and therefore more
effectively deterministic (the definition of RJ is unbounded/Gaussian to
least below 1E-12, and DJ is all other stuff).