Re: [STDS-802-3-EPOC] Structure of Cl 100 & 102
Duane,
Two monitor comments on this topic:
1) I believe you mean Clause 100 (PMD) and Clause 101 (PCS/RS) rather
than Clause 102 (MPCP), unless I am missing something altogether in this
discussion J
2) Regarding were to put PMA: if you look at it from spec organization
perspective, it does not really matter too much where you stick it - we like
complete layers to be housed in a separate clause for simplicity of reading
the material afterwards. However, functionally, it all depends on what kind
of implementations we expect to see in the future. Should we ever expect to
see (and allow) implementations in which PMD may be implemented separate
from the MAC/PCS, putting PMA together with PCS seems to make sense, given
that the PMA interface would then have to run between individual chips.
Should we be only looking for integrated solutions, where PMD is implemented
together with PCS, RS and PMA in a single slab of silicon, then it does not
really matter where PMA goes since all interfaces will be internal anyway,
and externally observable behavior will be exactly the same for PMA located
in Clause 101 or in Clause 100.
With that said, I am personally very much in favour of not restricting
implementations and allowing system designers work with both discrete
solutions as well as integrated ones. This seems to be the most flexible
path forward for EPoC.
I think the bigger question here should be rather where PMD ends in our
case, i.e., what kind of information should the PMD service interface
provide to the upper layers and what expect from them. If we can identify
clearly what we need each layer/sublayer to do, it is easy then to decide
where things go (i.e., to what clause PMA would belong to).
Regards
Marek
From: Duane Remein [mailto:Duane.Remein@xxxxxxxxxx]
Sent: Friday, 24 May, 2013 10:42 PM
To: STDS-802-3-EPOC@xxxxxxxxxxxxxxxxx
Subject: [STDS-802-3-EPOC] Structure of Cl 100 & 102
Your Editors have been discussing the structure of the draft, in particular
where to put the PMA layer material.
In most 802.3 clauses the split is RS/PSC & PMA in one clause and the PMD in
a separate clause. However this may be reflective of the Optical nature (or
perhaps more precisely the NRZ nature) of most 802.3 PHYs. There is also
much historical precedence for this structure and I would be remise I didn't
mention tendencies to split functionality into separate physical IC devices,
However, there is an exception to this; Clause 62, which is the only OFDM
based PHY currently standardized in 802.3, used a different structure. CL 62
addresses the PMA and PMD while Cl 61 covered the PCS (and TC layer, but
that is because so much material for this copper PHY was borrowed from ITU
specifications). I am wondering if this makes more sense for an OFDM(A)
based PHY given that the FEC is so closely tied to the modulation (IFFT/FFT)
scheme. It might simplify the interface descriptions for PCS/PMA and PMA/PMD
that we will need to define in our spec.
I would like some of our OFDM experts to comments on this topic. Certainly
everyone is invited to comment (in fact I've personally addressed a few long
standing 802.3 members who might have opinions on this topic) but I'm really
looking for technical reasons to go one way or another.
Best Regards,
Duane
FutureWei Technologies Inc.
duane.remein@xxxxxxxxxx
Director, Access R&D
919 418 4741
Raleigh, NC
_____
________________________________________________________________________
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