Dear Guo :
Thanks for your information collection on 25Gb/s APD receiver sensitivity , I see there is another contribution lee_3ca_2_0716 from Lee do the same things in the
meeting next week.
They are very useful for us to evaluate different solutions and define the parameters for 25G systems.
Just a different opinion on your following comments:
“This is actually an issue that where you choose to put more pressure,
transmitter side or receiver side. We think improving the sensitivity is much easier than increasing the transmitter power.”
Usually, launch power is easier to improve compared with sensitivity improvement, due to :
1)
The transmitter power from the laser chip is much easier to increase compared with APD chip sensitivity. For launch power, you can always increase the
bias current to get more power until the limit of saturation or damage , if you can improve the damage power of a laser chip and also improve the heat dissipation with assembling, then you can get more launch power. While for APD sensitivity, you don’t have
much measurement to improve the sensitivity except substantial technology improvement in chips fabrication.
2)
The current coupling efficiency of the transmitter is generally low, 30%~40% coupling efficiency is quite typical. There is still much space for you
to improve. If you can improve the coupling efficiency such as by more lens , or better package control, then you can improve it. While for a receiver , the coupling efficiency is very high already (more than 80%), there is no space(or very small) to improve.
3)
There is usually a 4~5dB launch power spread range (such as from 4dBm ~ 9dBm for PR30 U1/U3, so large spread range), so you can get higher launch power
just by “pick” (at some cost with yield loss). But for sensitivity , the sensitivity spread range is quite small, you don’t have any spacing to pick up.
The present case is that, if you ask a vendor a transmitter with higher launch power, it’s usually doable but just with higher price. But if you ask a vendor
a more sensitive receiver, the answer is usually “no” (not feasible in technical)
On the issue “where you choose to put more pressure, transmitter
side or receiver side”, I think a better answer is “the OLT side”, due to the OLT is less cost sensitive. That’s also a general principle in the past when we define the PON
spec.
But any way, look forward to see all of you and discuss the details next week in the meeting room.
Best regards
Dekun Liu
____________________________________________________
Advanced Access Technologies Dept.
网络研究接入技术部
Huawei Technologies Co., Ltd.
华为技术有限公司
Phone: +86 027-59267217 Email: liudekun@xxxxxxxxxx
湖北武汉市关山一路光谷软件园A7-9
邮编：430074
Huawei Technologies Co., Ltd.
A7-9 Wuhan Optical Valley Software Park,Guan Shan Road,Wuhan,Hubei, P.R.China
From: Yong Guo [mailto:guo.yong3@xxxxxxxxxx]
Sent: Friday, July 22, 2016 2:32 PM
To: STDS-802-3-NGEPON@xxxxxxxxxxxxxxxxx
Subject: [802.3_NGEPON] 答复: [802.3_NGEPON] Clarifying questions on guo_3ca_1_0716.pdf
Dear Ed,
Thank you very much for your comments,
For your clarifying questions, following are my quick answers.
1. Yes, in the table of page 5, minimum of 6dB ER is indicated. However, AVPmin, OMAmin and sensitivity are all based on ER=9dB. I'm just keeping the same style with the Table 75-5 in 802.3. Please correct me if I'm wrong.
2. Some of these contributions are listed as follows, I will update them as back up slides before the meeting. The proposed downstream sensitivity is based on 9dB ER.
cole_3ca_1_0316.pdf
pan_3ca_1b_0516.pdf
guo_3ca_1a_0516.pdf
tanaka_3ca_1_0516.pdf
3. Yes, actually all of our lab results are better than -28dBm. We also had test from our 25G-PON prototype, results show better than -27dBm. We believe performance could be further improved in the future. However, in contrast, we don't think increasing minimum
average power more than 3dBm for EML is feasible without using postamplifier. This is actually an issue that where you choose to put more pressure, transmitter side or receiver side. We think improving the sensitivity is much easier than increasing the transmitter
power. Larger output power with amplifier can be used to compensate WM insertion loss.
Thank you very much, and look forward to seeing you next week.
BR,
Yong
郭勇 Guo Yong
Ph.D
技术预研高级系统工程师
Senior Technology Preresearch System Engineer
FN产品团队
/ 有线产品经营部 FN Product Team / Wireline Product Operation Department
|
|
|
|
"Harstead, Ed (Nokia - US)" <ed.harstead@xxxxxxxxx>
2016-07-22 04:06
|
|
Yong Guo,
Thanks for your interesting contribution
guo_3ca_1_0716.pdf,
I do have a few clarifying questions:
On slide 5:
1. In the table ER=6 dB is indicated, while in the notes ER=9 is indicated. Could you please clarify? (I think you used 9 dB, and the 6 is a typo).
2. It is stated that “-27dBm sensitivity using 25G APD is feasible based on results from multiple contributions”. Could you please cite those contributions? Also,
could you please confirm that this is for ER=9 dB.
3. Are these lab results, or do you consider -27 dBm would be a viable commercial value for our standard (i.e. high yield, over temperature and end-of-life).
Thanks for your help.
Ed
Ed Harstead
Lead Technologist
Chief Technology Office, Fixed Networks, Nokia
+1 718 309 3967
ed.harstead@xxxxxxxxx
From: Glen Kramer [mailto:000006d1020766de-dmarc-request@xxxxxxxx]
Sent: Tuesday, July 19, 2016 3:02 PM
To: STDS-802-3-NGEPON@xxxxxxxxxxxxxxxxx
Subject: [802.3_NGEPON] contributions for the July meeting
Dear Colleagues,
We have 33 contributions to be considered at the meeting next week.
It will take me some time to create the html page with all the links. But the files are now available as a directory listing at the URL below, in case you want to start reviewing the presentations
sooner.
http://www.ieee802.org/3/ca/public/meeting_archive/2016/07/
Also, if you have submitted a file and don't see it uploaded, please let me know ASAP.
Regards,
Glen