|Thread Links||Date Links|
|Thread Prev||Thread Next||Thread Index||Date Prev||Date Next||Date Index|
Hi John and the rest of the group,
Here I am sharing with you part of the conversation I already had with John regarding the MPI calculation in 10GEPBud3_1_16a.xls.
In the Excel file, you enter RX/TX return loss values, the geometrical mean of the these are then calculated by the Excel and the "Upper bound method" by Vipul/Pete is used to calculate the eye closure. The results are then discounted by
- Channel Insertion loss considering the triple-reflection nature of RX/TX reflection
- Discount factor (currently set to 0.6) to represent the phase randomization effect
I have two basic reservations on this method
- The connector return-loss is not modeled. Probably the rational behind that was the Connector RL is much lower than RX/TX. However, this is not reasonable because (a) Unlike the RX/TX triple reflection, a reflection that includes the connector does not see the channel three times and it may go through IL as minimum as only once. (b) With 4 connectors, you get 14 extra reflections is addition to RX/TX. The cumulative effect of these cannot be ignored even with connectors return loss being low. You can scroll down in this email and look at my original calculations as an example.
- Applying a flat discount of 0.6 (or anything for that matter) to model phase randomization is probably not the correct way to do this. This is even worse if you add FEC (lets say to PSM4). Proper interaction of MPI with FEC should be analyzed.
On the positive side, I do agree with the following
- The phase randomization should not be ignored and should be modeled correctly. I am working on this and may share something with you comes May meeting.
- I recommend replacing the upper bound method with "Statistical Upper Bound" method presented a few month ago. This will give you some discount already.
Director System Eng
Date: Thu, 28 Feb 2013 17:26:32 -0800
Subject: Re: [802.3_100GNGOPTX] SMF Ad Hoc
Do you have results using the different calculation methods for 10GBASE-LR as defined in clause 52?
The draft minutes and presentation from today’s call are on the SMF Ad Hoc web page.
As discussed on today’s SMF Ad Hoc call, the next call of the SMF Ad Hoc will be at 8:00 am Pacific on Tuesday 18 December (1.5 hour duration if sufficient presentations otherwise shorter).
Please send requests for presentations on any of the SMF proposals to me.
There is a further SMF Ad Hoc opportunity on 8 Jan.
Peter Anslow from Ciena has invited you to join a meeting on the Web, using WebEx. Please join the meeting 5-10 minutes early so we may begin on time.
+44-203-4333547 (United Kingdom)
4438636577 (United States)
Pete Anslow | Senior Standards Advisor