| Thread Links | Date Links | ||||
|---|---|---|---|---|---|
| Thread Prev | Thread Next | Thread Index | Date Prev | Date Next | Date Index |
Hi, Eldad, Thanks. I actually disagree with you; those existing PMD sub-layers also include bit representation, channelization, on-air timing, etc. However, I’m not trying to argue with you about the value of making the 11ac (or 11af) PHY into a single structure without this sub-structure – the TG should make that expert decision. My statement was simply that the way the structure is described in clause 7, if you are going to delete one of the components, then you really don’t have any internal structure and are a single “monolith” known as PHY – and thus you should drop any usage of PMD or PLCP. Having said that, I do agree with you that the terminology usage in clauses 4, 6, 7 and the existing PHYs is not appropriate for such a single structure PHY, even though clause 7 implies that such structure is optional. This sounds like something to address in REVmc. For now, I’m withdrawing my comments/complaints about how to describe a PHY without such structure, and I’ll leave it at “just do the best you can – we’ll fix it in REVmc.” That’s my suggestion anyway. And, yes, there is irony (I’m being nice) in the Standard having internal structure that nobody actually uses, but be missing internal structure that is frequently used – you’re very accurate in that. Of course, it is only a logical description and not an implementation description and all that, so we can live with it. But, that seems like another topic to consider in REVmc, perhaps. I would also note that several years ago, ARC SC started working on a functional/component architectural model of the MAC, which just started to consider such actual usage of structure (in the sense of trying to help make sure our MAC model was enabling real implementation choices), but that work never completed. It may or may not be worth resurrecting. Mark From: Perahia, Eldad [mailto:eldad.perahia@xxxxxxxxx] Hi Mark, Brian and I reviewed the 11a and 11n PHY’s and as far as we can tell the only functionality contained in the PMD are things like RSSI/RCPI measurements. The preamble is considered part of the PPDU, which is described by the PLCP. So it would seem easier to keep the PPDU and PLCP terminology. One way to merge with Clause 4, 6, 7 layer diagrams is to have a single statement that for the purposes of TGac and TGaf the PMD is the “bottom” part of the PLCP. Then later in 11mc, the PMD concept can be excised from Clause 4,6,7 and that statement removed. It is interesting to note that while we split the PHY in to the PLCP and PMD in the spec, something that is not done in reality, that the MAC is a single entity in the spec. However people/implementations commonly consider the MAC in terms of upper and lower parts. Regards, Eldad From: *** 802.11 TGaf - TV White Spaces OperationTask Group *** [mailto:STDS-802-11-TGAF@xxxxxxxx] On Behalf Of Hamilton, Mark Similarly to my question in TGac on this, I am trying to understand what will be done instead (beyond just what is _not_ needed, that is). If the PMD concept is dropped, I do not think it makes any sense to have a PLCP concept. If TGac and TGaf would like to simply have a “PHY” (which delivers the PHY service as defined in clause7), that seems fine. I note, however, that clause 7 describes the PHY layer structure as having a PMD, and an optionally null PLCP which maps the PMD to the PHY service. So, I think you are actually suggesting to have a null PLCP. Mark P.S., my apologies to TGac for not making the discussion on this yesterday, as I had a conflict. From: *** 802.11 TGaf - TV White Spaces OperationTask Group *** [mailto:STDS-802-11-TGAF@xxxxxxxx] On Behalf Of Peter Ecclesine (pecclesi) Hi All, Adrian Stephens and I discussed these 11ac comments in the context of TGaf Clause 23, and it is my opinion that 23.6 PMD need not be specified in the TGaf PHY. petere From: *** 802.11 TGac - VHT below 6GHz*** [mailto:STDS-802-11-TGAC@xxxxxxxx] On Behalf Of Perahia, Eldad Hi all, There are two comments (CID 6442 and 6206) regarding cleaning up the PLCP-PMD interface. My opinion is that this interface is useless and the PMD should be deleted and the PLCP-PMD should be turned into just the PHY. I am willing to do such work if there is consensus with this position. I am not willing to do the work of cleaning up the interface. Today in the 802.11ac pre-meeting we had a group discussion on this topic (no submission). I proposed three alternatives: 1) reject the comments, 2) delete the PMD (I am willing to do this work), 3) transfer CIDs to someone else that is willing to do PMD clean-up work. The consensus of the group at the 802.11ac pre-meeting is to delete the PMD. No one spoke against deleting the PMD. We will have this discussion again next week. The 802.11ac chair has offered to schedule this discussion topic at the first TGac time slot. If you have an interest in this topic please attend or please email the reflector, especially if you are in favor of keeping the PMD. Regards, Eldad |