RE: [802.3ae_Serial] From serial PMD call 19 Mar: measurement plan for stressed sensitivity feasibility
For my simulations, I used a mathematically perfect 4th-order Bessel
Thomson filter. Actually two - 0.4x data rate and 0.75x data rate. The
edge rates going into the filters were essentially infinite. I checked
all this again this morning.
So, possible differences
1. Real filters have imperfections in phase and amplitude, particularly
that far out of band.
2. Other frequency limitations, reflections, etc. may exist in actual
setups.
3. Actual input edge rates are slower. In simulations, this has only
minor effect, as expected.
Tom
-----Original Message-----
From: piers_dawe@agilent.com [mailto:piers_dawe@agilent.com]
Sent: Friday, March 22, 2002 8:19 AM
To: stds-802-3-hssg-serialpmd@ieee.org
Subject: [802.3ae_Serial] From serial PMD call 19 Mar: measurement plan
for stressed sensitivity feasibility
The main output from this meeting was Stretch's updated email
http://www.ieee802.org/3/10G_study/public/serial_adhoc/email/msg00592.ht
ml .
Stretch pointed out that attempting to do the same thing in different
laboratories (as opposed to a round-robin approach with common parts)
has
merit because we learn something about consistency.
Discussion of filter bandwidth for stressed eye generator. Assume
filter
bandwidth is an electrical bandwidth and so is the 7.5 GHz standard
filter.
Can anyone confirm?
Petar's experience with different physical filters is that 5 GHz
electrical
is too much (too much ISI?) so how come Tom had 4 GHz?
We could use a common physical filter as input to both scope and error
detector - would have worse tolerance than the filter built into the
scope,
not sure about consistency.
Discussion of how to know if 1% of histogram is the right metric: sample
histograms would be interesting. Concern about fat tailed distributions
-
let's see.
It is hoped that each company will be able to measure at least a couple
of
DUTs, e.g. from two different batches.
Draft 4.2 expected soon.
Present
-------
Piers Dawe Agilent SPG
Peter Ohlen Optillion
Petar Pepeljugoski IBM
Stretch Camnitz Agilent T&M
David Kabal Picolight
Next phone meeting
------------------
Tuesday 26 March, at the usual time and coordinates:
4:15 pm GMT = 17:15 CET = 11:15 am EST = 8:15 am PST
+1(816)650-0631 Access code 39209
Piers