Robert,
It wasn't included in the spreadsheet, but
to generate the 25 cases for each type of pulse response, I swept my target
%tile from 88% to 99.5% in 0.5% increments, and included 99.8% to get close to
5dB as requested by Paul. The same %tile target was used for PIE-D and finite
DFE, i.e. I did not "relax" the finite DFE targets when looking for
symmetric and postcursor type responses. Of course, the %tiles are based on my
modeling assumptions, which I've presented previously. If these assumptions
change then the %tile interpretation will shift.
From:
owner-stds-802-3-10gmmf@IEEE.ORG [mailto:owner-stds-802-3-10gmmf@IEEE.ORG] On Behalf Of Lingle, Jr, Robert (Robert)
Sent: Thursday, April 21, 2005
11:24 PM
To:
STDS-802-3-10GMMF@LISTSERV.IEEE.ORG
Subject: Re: [10GMMF] Towards
receiver test responses
I may have missed it, but did you show
here what are the %tiles on Gen67YY w/r to the DFE cases and PIE-D?
Robert Lingle, Jr, Manager
Fiber Design and Transmission
Simulation
OFS R&D, Atlanta, GA
-----Original Message-----
From: John.Ewen@JDSU.COM
[mailto:John.Ewen@JDSU.COM]
Sent: Thursday, April 07, 2005
2:39 PM
To:
STDS-802-3-10GMMF@LISTSERV.IEEE.ORG
Subject: Re: [10GMMF] Towards
receiver test responses
The attached .xls file contains the candidate TP3 ISI
parameters that were requested during the TP3 telecon on 5-Apr-05. The PIE-D
targets were generally swept from ~3.5dB to ~5dB, although there will be some
variation in the resulting PIE-D due to the nature of the search algorithm.
The
results for the split-symmetric response deserve some comment. The first 11
entries in the worksheet are essentially identical. This occurs because
split-symmetric responses, as we have defined them, do not exist for PIE-D <
~ 4dB. For PIE-D targets < 4dB, the search algorithm finds the best response
possible consistent with the constraints resulting in the same ISI parameters
for these cases.
Regards
- John