Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

[EFM] [EFM-P2P] RE: EFM-P2MP WDM Overlays and possible modifications to the PAR or new working group



Gerry,

I'm not sure I completely understand your reply, the paragraph you wrote
on CWDM/DWDM/AnalogVideo, seems to fall back to the "out of scope"
statements heard all too often in my opinion, but at the same time you suggest
some solutions in your second paragraph.

You first paragraph reads:

"However, I doubt the CWDM/DWDM/AnalogVideo overlay wavelength grid and isolation requirements will be included in an 802.3ah specification. P2P and
P2MP EFM work will focus on an upstream and downstream wavelength for
Ethernet traffic.  Adding WDM on top of this for non-Ethernet services, like
Analog video or private xWDM Wavelength circuits, may be an important part
of an optical subscriber access system, but not defined in P802.3ah."

However, your next paragraph suggests several solutions, so I would like clarification on two points, one with regard to the first paragraph and the other for the second.

1. "a wavelength window and isolation for the transceiver" stated as "the
transceiver requires -40 dB isolation at 1550nm be sufficient" which means
the PONs would be compatible with 3-lambda transceivers, or in your words:
"to insure compatible 3-lambda transceivers"
I would not expect this 802.3ah working group to generate a detailed
specification for an analog video overlay signal. I'm OK with this, I'd personally
be very happy just to see us specify a wavelength window and isolation for the
transceiver as part of the 802.3ah specification.

2. I need clarification for your last sentence in that paragraph, when you say
"refer to the FSAN/ITU-T G.983.wdm specification for these overlay band
definition" I don't think you're saying that the 802.3ah specification would include
a recommendation to follow the FSAN/ITU-T G.983.wdm specification, right?
or are you ? In the January 2001 Irvine meeting I was shot down for even suggesting we adopt some sub-sections of the FSAN/ITU-T G983.*
specification. It seemed the "NIH" factor was high, I was pursuing this course so we would not have to reinvent/create many specifications which already exist and
work for P2MP access networks.

Your second paragraph reads:

"So is there a solution?   Kent - is this mostly an interoperability concern
for the analog video itself, or to insure compatible 3-lambda transceivers?
One requires detailed specifications of the overlay signal, and the other a
wavelength window and isolation for the transceiver.  In other words, would
something that just said the transceiver requires -40 dB isolation at 1550nm
be sufficient to address your requirement?  Or, is even that necessary;
why not refer to the FSAN/ITU-T G.983.wdm specification for these overlay
band definitions?"

Since I've gone this far in pursuing so called "out-of-scope" issues, why stop
now. For PON or FTTH to be successful it MUST carry voice, data, and video
of some type. We at minimum need to create specifications that defines
acceptable jitter, latency, etc..., to support these applications, without actually
creating specifications for the applications. I think that's how you told me one
would need to go about to create specifications for these types of applications.
If this can't be done due to our working group being tied to 802.3, we really need
to consider breaking off into a new working group such as 802.18 or what ever
the next number is. I heard this was brought up at a previous meeting, and failed,
but I must have missed it. I think using Ethernet in the First Mile is a great idea.
An Ethernet PON "ePON" can be built which supports the applications described
above, but it can't be done effectively unless areas/functions such as ranging,
security, DBA, PSS, PON frames to encapsulate Ethernet frames, and provide
OAM, are added to the MAC or a PON bridge chip is defined which would sit
between the PMD/SERDES and the MAC in the ONU and OLT.

Many have presented good solutions to many of these issues, but they are
always met with the same response, it's "out-of-scope." If ePON is ever to be
plug and play as Ethernet is today, then these issues must be in the PAR thus
in scope so a specification can be written, resulting in a standard. But that's just
my opinion, I could be wrong."-D.M." And I'm sure as this gets picked apart, many
will say I am. My goal here is for us to consider the big picture with respect to the
role PONs play in EFM.
 
All this said at the very minimum, I think we should be able to come up with a
specification for the PMD, Optical modules for P2P and of course P2MP.
 
 
Happy New Year!
 
Tony Anderson
tony@xxxxxxxx

In a message dated 12/29/01 11:50:37 PM Pacific Standard Time, gerry.pesavento@xxxxxxxxxxxx writes:

Subj: [EFM-P2P] RE: EFM-P2MP WDM Overlays
Date: 12/29/01 11:50:37 PM Pacific Standard Time
From:    gerry.pesavento@xxxxxxxxxxxx
Sender:    owner-stds-802-3-efm-p2p@majordomo.ieee.org
To:    stds-802-3-efm-p2mp@xxxxxxxxxxxxxxxxxx, stds-802-3-efm-p2p@xxxxxxxxxxxxxxxxxx


Tony wrote...
> At minimum, if we spec'd 1550 as a location for a video
> overlay we would need to define how many dB down, for
> isolation, a 1550 nm signal would be at 1490 (1480-1500), I
> think it needs to be in the neighborhood of -45db.

It is -40 dB.  It requires a good bandpass filter, and I am confident that
the transceiver manufacturers, if they want to offer a product that overlays
analog video on top of P2P or P2MP Ethernet, can do so.

Kent wrote...
> I would like to see the IEEE define a PON as 1310 data up,
> 1490 (1480-1500) data down and 1550 as video down

1550 nm is "it" for Analog video because of availability of lasers and
requirement for an EDFA to get through the splitter. 

However, I doubt the CWDM/DWDM/AnalogVideo overlay wavelength grid and isolation requirements will be included in an 802.3ah specification. P2P and
P2MP EFM work will focus on an upstream and downstream wavelength for
Ethernet traffic.  Adding WDM on top of this for non-Ethernet services, like
Analog video or private xWDM Wavelength circuits, may be an important part
of an optical subscriber access system, but not defined in P802.3ah.

So is there a solution?   Kent - is this mostly an interoperability concern
for the analog video itself, or to insure compatible 3-lambda transceivers?
One requires detailed specifications of the overlay signal, and the other a
wavelength window and isolation for the transceiver.  In other words, would
something that just said the transceiver requires -40 dB isolation at 1550
nm be sufficient to address your requirement?  Or, is even that necessary;
why not refer to the FSAN/ITU-T G.983.wdm specification for these overlay
band definitions? 

Happy new year - almost.

Gerry Pesavento
 
Tony Anderson
 
Dir. PON Technical Applications
ZONU