[EFM] Minutes from PON Optics Telephone Conference - 19th December
Attendees
----------------------------------
Frank Effenberger
Gerry Pasavento
Glen Kramer
Lior Khermosh
Mike Wirtz
Piers Dawe
Tom Murphy
Tremmont Mial
PON TIMING VALUES
------------------------------------
No further discussion of this issue
SIGNAL DETECT
------------------------------------
We agreed upon values for the Signal Detect Tables
For all four tables the entry Input_Optical Power <= -45 dBm
The Off Power of the OLT and ONU Tx to be set to -45 dBm
These changes will be submitted by me in the form of comments to D1.2
What the SD in the upstream actually does is not yet clear. Three
possibilities were outlined:
1) Fast SD which tracks individual bursts and can be used to identify an ONU
drop-out
2) Slow SD which is a very loose envelope over bursts from several ONUs
where the function
is to prevent the OLF from trying to lock into noise, its own signal
etc.
3) No SD and the PCS somehow deals with the non-defined 'stuff' it becomes
between
bursts, or it receives some signalling from higher layers which can act
as SD:
The feeling of people on the call was that Option 1 is overkill as this
functionality
can be better provided at a higher layer. Option two seems the easiest to
implement and most
sensible.
Two important questions need to be answered:
1) What is the shortest burst length. This is a orientation point in
deciding if SD can/must
react in a single burst
2) Do we need changes to the PCS which is expecting SD to synchronise
8B/10B coding.
Next Steps
------------------------------------
Feedback from PCS and protocol people regarding the above two points
Submission of comments and subsequent discussion at the Vancouver meeting
Hopefully more e-mail discussion timing before Jan.
Next Meeting
----------------------------------
I am not sure if we will have enough people for a meeting for next Thursday.
Please mail if you are available.
All the best
Tom