Updated link model spreadsheet for 802.3ae

["DAWE,PIERS (A-England,ex1)" <piers_dawe@agilent.com>]

Dear all,

Please find on the IEEE web site, a revised link model spreadsheet.  It's at
http://www.ieee802.org/3/10G_study/public/email_attach/3pmd046.xls .

This spreadsheet has updated equations as compared with the latest on the
IEEE server which is
http://grouper.ieee.org/groups/802/3/10G_study/public/email_attach/All_1250v
2.xls .

Some of you have seen and used draft spreadsheets such as 3pmd041.xls and
3pmd043.xls.  There are no equation changes compared with those, so for you,
nothing has changed.

For everybody's convenience, the spreadsheet is populated with a set of
numbers representing three "LAN PHY" PMDs as described by Del Hanson
http://www.ieee802.org/3/ae/public/may00/hanson_1_0500.pdf .  However, your
approval or not of these PMDs is a separate issue.  I expect we will be
voting on the equations, structure, format of the model in one vote, and the
PMDs in other vote(s).

As compared with the version presently shown on our web site,
http://www.ieee802.org/3/10G_study/public/email_attach/All_1250v2.xls , the
revised equations here:
1.	Bring the receiver bandwidth-risetime calculation into line with the
rest of the spreadsheet,
2.	Make the model a little less pessimistic for extremely high ISI
penalties,
3.	Calculate TP4 eye opening penalty on the same basis as other ISI.

Please find at
http://www.ieee802.org/3/10G_study/public/email_attach/new_isi.pdf a
document by Dave Dolfi which explains the changes 1 and 2 (the Greek letters
sigma and pi have come out as s and p - sorry!).  Change 3 uses a
generalisation of the same equations that Dave describes.  If interested,
you can see what's going on by analogy to the revised ISI calculations and
inspecting the spreadsheet.  I hope to make a presentation on this next
week.

You may type in the Gigabit Ethernet values and compare with All_1250v2.xls
or the classic IEEE 802.3z link model spreadsheet,
http://www.ieee802.org/3/10G_study/public/email_attach/All_1250.xls .
You'll see there's not much difference in outcome.  However, the updated
equations provide a self-consistent framework for any model extensions
needed moving forward.  For an explanation of the differences between
All_1250.xls and All_1250v2.xls, see
http://www.ieee802.org/3/ae/public/mar00/dawe_1_0300.pdf .

To change values in this spreadsheet you'll need the Error function "Erf"
loaded into Excel.  You need do the following only once, or it may be
already loaded:
a. Under the Tools menu at the top of the spreadsheet, select  "Add-ins".
b. Under the Add-ins pop-up menu, choose the "Analysis Tool Pack"
c. Click "OK"
d. Close Excel and re-open it.  The error function should now be enabled.

See Dave's document, appendix, for more info on this if needed.  We believe
this spreadsheet works under Excel 97 and Excel 2000 on PCs.

I hope that those who need it will find a useful and stable tool here.
Enjoy!

Piers

    Phone   +44 (0)1473 465654  or   Telnet (31)25654
    Fax	  +44 (0)1473 464965   piers_dawe@agilent.com
    Agilent Technologies   White House Road   Ipswich
    Suffolk    IP1 5PB    England