Re: [8023-10GEPON] 10GEPON ad hoc on high split systems - Task 1 - corrected link
Marek,
Several people have commented to me that including company names and links
to their data sheets, even if these are publicly available, can be
considered an endorsement/advertisement. So, in order not to avoid any
possibility of a complaint, please do not include the links. I still would
recommend that you add the table with the insertion loss numbers from
different manufacturers, but do the following:
a) do not include links to company's web sites
b) replace Company names with "Vendor A", "Vendor B", etc.
Thank you,
Glen
_____
From: Hajduczenia, Marek [mailto:marek.hajduczenia@siemens.com]
Sent: Wednesday, October 25, 2006 10:52 AM
To: glen.kramer@TEKNOVUS.COM; STDS-802-3-10GEPON@listserv.ieee.org
Subject: ODP: [8023-10GEPON] 10GEPON ad hoc on high split systems - Task 1 -
corrected link
Glen,
thank You for the suggestions. I added a slide and You can find an updated
version of the presentation at:
http://www.ieee802.org/3/av/working_docs/Adhoc-task1.1.ppt
The included links to manufacturers' data sheets regarding PSC modules are
meant as an example only rather than an exhaustive list. The quoted data
sheets are open to public and even though I had access to some confidential
material, I did not included it in the estimation and the final results.
That would be a violation of the IEEE policy for open information access.
Regarding the methodology of the research. I should probably clarify a
little bit the way data was collected and processed. Please have a look at
the file: <http://www.ieee802.org/3/av/working_docs/Adhoc-task1.1.ppt>
http://www.ieee802.org/3/av/working_docs/
<http://www.ieee802.org/3/av/working_docs/Splitter_loss1.1.xls>
Splitter_loss1.1.xls
The data sheets were first collected from individual manufacturers and then
divided into FBT and PLC devices. Since the manufacturing technologies are
different, it made sense to differentiate the examination as well, since the
characteristic insertion loss for both aforementioned technologies are
slightly varied. The data from the data sheet was next collected in the tab
entitled "Companies". Each entry has its PSC module description indicating
whether the given device is dual window, PLC/FBT, and what quality is it.
Please note that the insertion loss typically decreases along with the
increase in the class of the device, but so does the cost. Next, the FBT and
PLC splitters were examined in more detail in tab "Splitter comparison",
where individual port counts had their average, min and max insertion loss
estimated for the available devices. Next, the excess loss (difference
between the measured insertion loss and theoretical insertion loss) was
calculated for average, min and max insertion loss estimated for the
available devices and they were plotted against the port count. The obtained
curves were approximated with log plots and approximation coefficients were
taken down. The said coefficients were then used to approximate the excess
loss for 1x64 and 1x128 port devices, since such high port densities are not
available commercially (apart from company manufacturing 1x64 PLC devices)
and then total insertion loss, which is a total of theoretical loss and
excess loss.
Having 1x64 and 1x128 port device insertion loss, we can estimate the total
link loss budget for 10 and 20 km systems, with 1x64 and 1x128 port
splitters in a single and multistage configurations. Obviously, a single
stage configuration features only 1x64 port deivce, while dual stage
configuration may have 1x8 + 1x8 or 1x16 + 1x8 etc. Several possible
configurations are exminaed in the said Excel file in the tab named "Power
budget". Two separate scenarios were examined in there, namely a
connectorized splitter (a splitter is connected to the fibre cable using
connectors) and a spliced splitter (a splitter is spliced with the fibre
plant). The fibre loss of 0.5@dB/km includes cable loss, splice loss etc and
is based on the values adopted in 802.3 - 2005, table 60.1. Probably we
should again look at this value since we will be operating @10G, though I
would not expect any dramatic changes.
If there are any questions regarding the methodology, any comments on the
values and obtained link loss budgets, please let me know. I want to be able
to improve this presentation but I need feedback from You on this.
Thank You for Your time
Marek
_____
Od: Glen Kramer [mailto:glen.kramer@TEKNOVUS.COM]
Wysłano: Śr 10/25/2006 5:31
Do: STDS-802-3-10GEPON@listserv.ieee.org
Temat: Re: [8023-10GEPON] 10GEPON ad hoc on high split systems - Task 1 -
corrected link
Marek,
Thank you for posting this. I suggest you add one more slide, where you list
links to the manufacturer's data sheets. You already have those links in
your excel file.
Glen
> -----Original Message-----
> From: Hajduczenia, Marek [mailto:marek.hajduczenia@SIEMENS.COM]
> Sent: Tuesday, October 24, 2006 9:53 PM
> To: STDS-802-3-10GEPON@LISTSERV.IEEE.ORG
> Subject: [8023-10GEPON] 10GEPON ad hoc on high split systems - Task 1 -
> corrected link
> Importance: High
>
> Dear all,
> seems that the spaces in the link which I posted previously were converted
> into some non-sense ...
> the link displayed as: "http://www.ieee802.org/3/av/working_docs/Ad
> <http://www.ieee802.org/3/av/working_docs/Ad> hoc - task 1.ppt"
> should be read as: http://www.ieee802.org/3/av/working_docs/Adhoc-
> task1.ppt. This is a correct link. Sorry for the problem.
> Best wishes
> Marek
>