Hi David,
Thanks for your comments.
I would like to suggest that you prepare a specific presentation for the September meeting.
If you want to change the stressed receiver sensitivity you will need to make a detailed proposal that will replace it.
I think that we first need to have a debate how to specify a PAM4 signal.
For the benefit of this discussion I added some comments below in “red”.
Kind regards,
Peter
Peter Stassar, 施笪安
Technical Director, 技术总监
Huawei Technologies Ltd, 华为技术有限公司
European Research Center, 欧洲研究所
Karspeldreef 4, 1101CJ Amsterdam
The Netherlands
Tel: +31 20 4300 832
Mob: +31 6 21146286
Hello Peter,
Unfortunately I will not be able to join the ad-hoc call on Tuesday so I would like to comment on your presentation now.
I agree that the new clauses require definitions for PAM-4 OMA, both outer and inner. However I don’t support the idea of specifying maximum receive power or stressed receiver sensitivity in terms of OMAinner. My reasons are as follows:
1. I think we can define an inner eye inequality parameter that provides the ratio of the smallest inner eye to the outer eye. We could then have 3 test signals where the upper, mid, and lower inner eyes were the smallest in each case.
Peter: I think that would be a transmitter specification and not so much a receiver specification.
2. The way receive power is measured in a test system is with either a wide area average power meter or with an oscilloscope that is calibrated by reference to a wide area average power meter.
3. In the case of the oscilloscope, when measuring a low-level optical signal as part of a sensitivity test, oscilloscope noise is a significant factor. For an NRZ or a PAM signal, if we know the extinction ratio, we can derive OMAouter from the average power measurement but to measure OMAinner at the receiver input we would need a very sensitive oscilloscope (better than the DUT).
4. The error detector used for the sensitivity test provides us with a count of errors after the PAM decoder but not a separate error count for each of the 3 inner eyes in the PAM signal. It might be possible to construct separate error counts using additional processing on the bit errors in the decoded pattern. However I don’t see the benefit of doing that for a PMD sensitivity test, although it might be useful for troubleshooting a design.
So, my proposal for the receive power and stressed receiver sensitivity parameters is that we specify 3 sets of unequal inner eyes and define overload power and stressed sensitivity with the worst case of the 3 signals.
Regards,
David Lewis
Lumentum
Dear Task Force colleagues,
While working the SMF clauses 122 and 123 I have come across some topics that are missing from the adopted Baselines and some others that also require inputs.
I have prepared the attached presentation for the next SMF Ad Hoc on Tuesday 1 September, which I would already like to share with you in order to get a good discussion at this Ad Hoc.
In the case that you may have questions or remarks, please feel free to contact me.
Thanks in advance for your inputs.
Kind regards,
Peter
Peter Stassar, 施笪安
Technical Director, 技术总监
Huawei Technologies Ltd, 华为技术有限公司
European Research Center, 欧洲研究所
Karspeldreef 4, 1101CJ Amsterdam
The Netherlands
Tel: +31 20 4300 832
Mob: +31 6 21146286
Dear Task Force Participants,
This email is to announce the preview draft of the IEEE P802.3bs 400GbE standard will be available on Friday, September 4. It will be posted in the private area.
My thanks to Mr. Anslow and the editorial team for all of their work on pulling this draft together.
Regards,
John D’Ambrosia
Chair, IEEE P802.3bs 400GbE Task Force