Re: [8023-10GEPON] Wavelength plan proposal : current status
Marek, Tanaka-san,
Regarding case-2, the filter does not need to have a notch... it is a pass
band filter... I think the chart on page 4 has the filter curve drawn upside
down. The whole point of that style of filtering is that you use a simple
pass-band filter in front of the detector.
If we have problems making passband filters, then we have big problems.
Regarding the suggestion of using only 6nm of passband for the new
transmitter, that will force cooling. Perhaps cooling is a foregone
conclusion at this point, but still, it will make hardened OLTs nearly
impossible. If possible, a window of 15nm would be much better.
Regards,
Frank E.
p.s. I didn't receive the original Email from the exploder... was that just
me?
-----Original Message-----
From: Hajduczenia, Marek [mailto:marek.hajduczenia@SIEMENS.COM]
Sent: Tuesday, May 08, 2007 7:21 AM
To: STDS-802-3-10GEPON@listserv.ieee.org
Subject: Re: [8023-10GEPON] Wavelength plan proposal : current status
Dear Tanaka-san,
Thank You for the splendid presentation and high quality report on the
progress of this particular ad hoc.
IMHO, I would not like to see two wavelength channels standardized for
downstream channel for a single reason: even though it is relatively easy to
manufacture ONU modules with different wavelengths (change of complete
optical subassembly with laser, filters and receiver), it may cause stock
problems and accidental mix-up when a system set up to operate with 10G
downstream in 1480-1500 nm window suddenly has an ONU which expects a
downstream channel in the 1574-1580 nm band. The OLT would have to be also
equipped with varying laser sources.
As much as it is a practical solution, I do not know whether it is the best
choice possible. I would rather stick with a single downstream band for the
sake of simplicity and transparency of the system.
As for the proposal of the filter structure in the presentation - I must say
that the case-2 (see page 4 for example), will require pass band filter of
quite super quality, which will have to be uniform for a large band
(upstream 1310 window, downstream 1490 window, video 1550 window), then have
a notch above 1570 nm and then again have minimum loss in the window above.
I am concerned whether such devices can be designed in practice. At least
what I gather from the work I did with Joao Santos, it is quite challenging
to have a filter with a large flat top, then sharp, high isolation narrow
band.
This is mainly a practical concern - I do not know how such a structure will
impact the other channels i.e. what will happen when the part of the video
channel leaks into the 10G signal.
Additionally, I did some minor editorial changes in the presentation. You
can find it attached. In its current form, I am ready to support its
conclusions. It would be beneficial also to add a slide on the proposed 2
option wavelength allocation system and consider its pros and cons. The
topic came out already so we may as well tackle it in the right ad hoc
group.
Best wishes and thank You for Your time
Marek Hajduczenia (141238)
NOKIA SIEMENS Networks S.A., Portugal - R
Rua Irmãos Siemens, 1
Ed. 1, Piso 1
Alfragide
2720-093 Amadora
Portugal
* Marek.Hajduczenia@siemens.com
http://www.marekhajduczenia.info/index.php
(+351.21.416.7472 4+351.21.424.2082
-----Original Message-----
From: Keiji Tanaka [mailto:kj-tanaka@KDDILABS.JP]
Sent: quarta-feira, 2 de Maio de 2007 5:42
To: STDS-802-3-10GEPON@listserv.ieee.org
Subject: [8023-10GEPON] Wavelength plan proposal : current status
Dear Colleagues,
Attached please find the intermediate report on wavelength plan proposal,
which is based on local discussion in Japan. In addition, I received study
results from Dr. Joao Miguel Santos of Siemens and discussed the filter
configuration issues. I appreciate his contribution.
I would like to elaborate the wavelength plan from the proposal in the
slides, and it would be appreciated if you could give me any comments and
technical proposals. I wish to apologize in advance for the
inconvenience that I will be out of my office from May 2 to 8.
Through the study, I recognize the difficulty of deciding the downstream
wavelength. The situation of broadband services and its access technologies
varies widely depending on countries and operators. Some operators may not
care the coexistence among 1G-, 10G-EPON, and RF-video. As a solution to
resolve the problem, is it possible to specify two wavelenth band for
downstream? For example,
Option-1: 1480-1500 for greenfield deployment,
Option-2: 1574-1580 for the coexistence among 1G-, 10G-EPON and RF-video.
I hope that the concept of creating options is acceptable.
Best regards,
Keiji Tanaka