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

[STDS-802-3-400G] FW: Comparison of SMF experimental data



I am resending my reply to Takai’s comments which originally got rejected by the reflector, due to some confidential text at the end of Takai’s original email.

Gary 

From: Gary Nicholl <gnicholl@xxxxxxxxx>
Date: Wednesday, May 20, 2015 at 8:59 AM
To: 高井厚志 <atsushi.takai@xxxxxxxxxx>, "STDS-802-3-400G@xxxxxxxxxxxxxxxxx" <STDS-802-3-400G@xxxxxxxxxxxxxxxxx>
Subject: Re: Comparison of SMF experimental data

Takai,

Thanks for your comments. 

A couple of points:

  • With regard to Rx sensitivity for NRZ I did use the -8.8dBm number presented in shirao_3bs_01a_0515  at TP2 and converted this to what is seen at the PIN using 3dB for the demux loss (as specified in the table). The column for “Rx Baseline Unstressed Rx Sensitivity Inner OMA  (dBm)” is the sensitivity at the PIN and not the sensitivity at TP3 (which is what goes in the standard). The reason I did this is that all of the measured results presented were for single channel, and therefore after any demux.
  • I did take the measured data from takai_3bs_01a_0515. I used the data in slide 6 (-12dBm @ 2e-4) for RX1 and the data in slide 8 (-14dBm @ 2e-4) for RX2. If I have missed something here, please accept my apologies and I will gladly make the change. It was a lot of work pulling together all of this data and trying to put it into a format that I could understand, and it is quite possible that I may have missed something. 
  • I would also point out that Alan was unable to apply his correction factor to the NRZ data as none of the receiver parameters used for the measurements were presented (as far as I could see ?).
  • One other thing I should point out is that this analysis was simply based comparing the  Rx receiver sensitivity, as this appeared to be one point where there was a lot of discussion on technical feasibility. I was not trying to say this is the only point of discussion. In hindsight I should have perhaps made this a little clearer. There is obviously a trade off between margin on the Tx and Rx. You can see this on the 4x100G data, where there should be ~ 5dB more margin for 500m compared to 2km (1 dB less loss margin and 4dB for the removal of the mux/demux).  However  if you look at the 4x100G 500m results in the table it is obvious that a significant portion of this link margin (~ 3dB) has been applied to providing more margin on the Tx (not specifically addressed in this presentation).

Thanks again for your comments, and again apologies if I miss-reported some of your  (or anyone else’s) data. I am just a simple system guy trying to understand all the data in a format that enables me to follow the discussions and hopefully make a rational, engineering judgment on the best path forward.

Gary 



From: 高井厚志 <atsushi.takai@xxxxxxxxxx>
Date: Tuesday, May 19, 2015 at 5:58 PM
To: 高井厚志 <atsushi.takai@xxxxxxxxxx>, Gary Nicholl <gnicholl@xxxxxxxxx>, "STDS-802-3-400G@xxxxxxxxxxxxxxxxx" <STDS-802-3-400G@xxxxxxxxxxxxxxxxx>
Subject: RE: Comparison of SMF experimental data

Gary

 

Thanks.

My comments

(1)   Your comment for 10km NRZ in page 6 should be changed.
The sensitivity is specified with reasonable margin using obtained data.
We specified the sensitivity of -8.8dBm.
Do you think NRZ need more margin?

(2)   You should list my data of -17.6dBm given in takai_3bs_01_0515.
Then the margin you listed will be 2.8dB.

 

The margin for sensitivity is depends on transmission scheme.

PAM4 may need more margin than NRZ.

 

Atsushi Takai

Oclaro Japan, Inc

 

 

 

From: Gary Nicholl (gnicholl) [mailto:gnicholl@xxxxxxxxx]
Sent: Tuesday, May 19, 2015 11:43 AM
To: STDS-802-3-400G@xxxxxxxxxxxxxxxxx
Subject: [STDS-802-3-400G] Comparison of SMF experimental data

 

While preparing for the Pittsburg meeting last week, I became very frustrated in trying to compare all of the experimental data presented in support of the differing SMF proposals. 

 

The results were often presented using different receiver  parameters (average power, outer eye OMA, inner eye OMA, etc) and with different receiver implementations/specifications.

 

This made it extremely difficult to easily compare the different experimental results.

 

I decided to take all of the experimental data presented in the  task force, capture  it in a single spreadsheet, and convert everything to inner eye OMA sensitivity (and at a 2e-4 BER) to make it easier to compare.  

 

As a second step I applied a correction  factor to all of the measurement data (where possible) to enable a comparison based on a common set of receiver specifications that are projected to be available in  realizable products for  25Gbaud and 50Gbaud systems in the time frame of the 802.3bs project. 

 

I shared this analysis with several people yesterday and they suggested it would be worthwhile sharing it with the task force. 

 

As a result I captured my analysis in the attached presentation.

 

Gary