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

Re: [STDS-802-3-EPOC] PHY Link ad hoc call tomorrow (Wednesday) 10am Pacific



Bill, 

 

Your question about the PHY delay needs further details before it can be
really answered. 

 

First, we need to know where this said timestamp is inserted in the stack
(before FEC encoding, after FEC encoding, below or above the MAC, etc.). 

 

Second, if we go with the clock different than even multiple of MPCP clock
(TQ units), we will have jitter to deal with - it is only unavoidable when
the base frequencies are off. 

 

Third, if we settle on sending PHY frequency reference information (PHY
timestamp) over the link, it is not clear to me how to synchronize that with
the MPCP clock, given that both of them are separated through the MAC and
XGMII, which do not carry any "synchronization" lanes to deal with aligning
both time domains. Said in different words, how can MPCP layer be aware of
what the PHY clock domain is at the given point of time to properly do
calculations on the other end of the link? If both clocking domains are
independent from each other, that would solve the problem, but the question
then raised on what we are really trying to achieve here - just sync the
PHYs together or provide clock transport to upper layers ?

 

Regards

 

Marek

 

From: Powell, William E (Bill) [mailto:bill.powell@xxxxxxxxxxxxxxxxxx] 
Sent: Wednesday, July 31, 2013 2:27 AM
To: STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx
Subject: Re: [STDS-802-3-EPOC] PHY Link ad hoc call tomorrow (Wednesday)
10am Pacific

 

Hi Mark, (or Avi)

Will Avi be on the call tomorrow? 

 

 I'm particularly interested in how his proposal for the PHY/PLC  timestamp
relates to the MAC MPCP counter, and whether the time delay through the OFDM
PHY is constant over multiple power up/down cycles (synchronized to the MAC
MPCP counters?).

 

Regards,

Bill

 

Begin forwarded message:

From: Mark Laubach <laubach@xxxxxxxxxxxx>
Date: July 30, 2013, 7:55:37 PM EDT
To: <STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx>
Subject: Re: [STDS-802-3-EPOC] PHY Link ad hoc call tomorrow (Wednesday)
10am Pacific
Reply-To: Mark Laubach <laubach@xxxxxxxxxxxx>

Fine by me.

 

From: Duane Remein [mailto:Duane.Remein@xxxxxxxxxx] 
Sent: Tuesday, July 30, 2013 4:47 PM
To: Mark Laubach; STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx
Cc: Duane Remein
Subject: RE: PHY Link ad hoc call tomorrow (Wednesday) 10am Pacific

 

Mark,

Could we have an ope discussion on the following topics?

Timestamp - this was part of  Avis presentation that was approved by motion
but there were no details. I would like to propose that this is a 32 bit
counter clocked off the PHY Clock (204.8 MHz) or some multiple thereof. The
CNU would keep a local copy of this clock (reset on receipt of DS PHY Link
timestamp) with behavior very similar to the local clock in EPON. This would
be part of the PHY frame overhead along with the already approved PHY
Configuration (2b), PHY Target Address (48b) and FEC Pointer (here I recall
14b but haven't found it in any motioned material).

 

Another item from Avi's presentation would be Start and number of RB's for
initial Ranging window. As a starting point here I would propose a 32 bit
counter clocked off the PHY Clock (204.8 MHz) or some multiple  thereof for
the start timer, call it PHY Discovery Start (essentially the same size as
the Timestamp). When the local PHY clock = PHY Discovery Start then the PHY
ranging window opens. A second register, call it PHY Discovery Duration (in
symbols or PHY clocks), would signify the end of the window.

 

Avi's presentation also included references to the Preamble sequence. I
would like to here more on what this is to include.

 

A detailed discussion on PHY Discovery would be good to have also, it should
include:

.         What information must the CNU acquire prior to responding to the
PHY Discovery window? My guess: FDD/TDD flag, FFT Size, Cyclic Prefix size,
Window Size, Interleaver Depth (?), OFDMA Frame size, US PLC Center
Frequency, US PLC Cycle Time. Anything else absolutely required?

.         What conditions on the CNU must be met on the CNU prior to
responding to the PHY Discovery window? My guess is only that the local PHY
clock and counter are synchronized.

.         How does the CNU respond and with what information? My guess is it
only needs to transmit the preamble sequence the its MAC address.

After this if there is time (ha ha) we could start on the "fine tuning".

               

 

Best Regards,

Duane

 

FutureWei Technologies Inc.

duane.remein@xxxxxxxxxx

Director, Access R&D

919 418 4741

Raleigh, NC

 

From: Mark Laubach [mailto:laubach@xxxxxxxxxxxx] 
Sent: Tuesday, July 30, 2013 7:02 PM
To: STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx
Subject: [STDS-802-3-EPOC] PHY Link ad hoc call tomorrow (Wednesday) 10am
Pacific

 

Dear IEEE P802.3bn Task Force participants,

 

There will be a PHY Link ad hoc conference call starting at its new time of
10AM Pacific time tomorrow (Wednesday).

As of this time, I have received no presentations to socialize for
tomorrow's call.   

We can use the time to review work plan and discussion.

 

Please be familiar with the IEEE Patent Policy:

https://development.standards.ieee.org/myproject/Public/mytools/mob/slideset
.pdf

 

Conference Call URL:

WebEx Meeting Number: 921 130 644

To start: https://broadcom.webex.com/broadcom/j.php?J=921130644

Meeting Password: This meeting does not require a password.

 

Yours truly,

Mark Laubach, Chair

IEEE P802.3bn Task Force

 

Broadband Communications Group

Broadcom Corporation

1351 Redwood Way

Petaluma, CA, 94954

  broadcom.jpg

Tel: +1.707.792.9093

Cell: +1.650.996.2219

 

 

  _____  

<="" p="">

 

  _____  

 

<="" p=""> 

 

  _____  


________________________________________________________________________

To unsubscribe from the STDS-802-3-EPOC list, click the following link:
https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-3-EPOC&A=1

JPEG image