Re: [STDS-802-11-TGM] Oh, *that* energy detect?!

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

In line, below….

Mark

-----Original Message-----
From: Mark Rison [mailto:m.rison@xxxxxxxxxxx]
Sent: Monday, November 19, 2012 11:16 AM
To: Hamilton, Mark
Cc: STDS-802-11-TGM@xxxxxxxx
Subject: RE: Oh, *that* energy detect?!

Hello Mark,

OK, thanks, so bonus question, arguably not in scope of this group, but anyway...

If CCA-ED is only for the 3 GHz band in the USA (a.k.a. 11y), why does

11ac/D4.0 have something about it (22.3.19.5.2 CCA sensitivity for operating classes requiring CCA-ED)?

[MAH] Don’t know. Presumably an oversight that they wouldn’t need it. Or, perhaps, provided just in case 11ac is ever extended to the 3 GHz band?

Actually one extra point of clarification:

> [CS/CCA and CCA-ED discussion] all this just feeds into the overall CS

> mechanism, by combining with NAV, TX status and PHY-CCA

I thought PHY-CCA (7.3.5.11) was the output of the thing which combines CS/CCA and (where appropriate) CCA-ED? Are you saying there is a different PHY-CCA which is completely separate from CS/CCA and CCA-ED?

[MAH] Sorry, poor wording on my part. I meant “by combining NAV, TX status and PHY-CCA” (without the “with”). Yes, PHY-CCA is the combination of CS/CCA and CCA-ED. Then, the MAC combines this with NAV and TX state. Sorry about the confusion.

Mark

P.S.:

> I'd say yes to all the above, except I think you meant the _fifth_

> being "the behavio(u)r ..."

Yup, sorry!

Mark RISON, Systems Architect, Wi-Fi English/Esperanto/Français

Samsung Cambridge Solution Centre Tel: +44 1223 434600

Innovation Park, Cambridge CB4 0ZT Fax: +44 1223 TBCTBC

ROYAUME UNI WWW: http://www.samsung.com/uk

> -----Original Message-----

> From: Hamilton, Mark [mailto:Mark.Hamilton@xxxxxxxxxxx]

> Sent: 17 November 2012 05:18

> To: m.rison@xxxxxxxxxxx

> Cc: STDS-802-11-TGM@xxxxxxxx

> Subject: RE: Oh, *that* energy detect?!

> Mark,

> That all sounds correct, yes. Note that the term "CS/CCA" is new as of OFDM (it was just called "CCA"

> before that, but was clearly a CS and/or ED mechanism then, per the

> "CCA Mode x" tables I showed you this week). And, yes, CCA-ED is a separate thing, optional and required per regulatory domain rules.

> If you think you can simplify it, I'll look forward to your proposal.

> Don’t forget that all this just feeds into the overall CS mechanism,

> by combining with NAV, TX status and PHY-CCA, so there is more mess within the MAC, too.

> CCA-ED is discussed explicitly in DSSS, HR/DSSS and OFDM PHYs. PHYs

> after that do, in my opinion, talk about still requiring the earlier

> clause PHYs' CCA pretty clearly. So, yes, I think by implication, CCA-ED is expected for all PHYs.

> I would have said that CCA-ED threshold is in a MIB variable for 'future/unforeseen' flexibility.

> Like, a new reg domain comes along that needs a different (lower

> power?) level, so an external entity can force this, while the Standard catches up. Since the hard-coded level is in Annex D, it seems

> clear that this is "subject to change" if the regulations change, also. (In fact, there was

> discussion a few months ago about moving all the operating class

> parameters into ANA, so they could change rapidly/frequently without

> needing to amend the Standard. I'm not sure what happened to that

> idea.) I have no idea about the other ones, except maybe someone

> decided all the above should apply to all the thresholds, but too late

> to really fix everything up cleanly. This all does sound ripe for a D1.0 comment (including the ANA part).

> > - What's the difference between "CCA-ED", "CCA-EnergyDetect",

> > "CCA-Energy Detect", "CCA energy

> detect" and "CCA-EDBehavior"?

> > Are the second and fourth just typos for the third, and the first

> > just an abbreviation of the third;

> the fourth being the behavio(u)r associated

> > with regdoms which require CCA-ED?

> I'd say yes to all the above, except I think you meant the _fifth_

> being "the behavio(u)r ..." I would also clarify that the fifth is

> the keyword for use in the Annex D table, only, also - so it isn't the

> behavior generically (that is CCA-Energy Detect), it is the keyword that tells you which domains require that behavior.

> Mark

> -----Original Message-----

> From: Mark Rison [mailto:m.rison@xxxxxxxxxxx]

> Sent: Friday, November 16, 2012 1:48 PM

> To: Hamilton, Mark

> Cc: STDS-802-11-TGM@xxxxxxxx

> Subject: Oh, *that* energy detect?!

> Hello Mark,

> So let me see if I've now grokked this:

> - CS/CCA includes both a "sniff the PHY header" mode and a "do energy

> detect" mode. The latter asserts CCA if the power is at least -62 dBm

> (assuming 20 MHz). These modes are both mandatory

> - There's a *separate* "do energy detect" mode called CCA-ED which is

> the same as the "do energy detect" mode of CS/CCA except the threshold

> is -72 dBm. This mode is only used in some regdoms, namely the 3 GHz

> band in the USA (a.k.a. 11y)

> If so, this is a very confusing way to describe it for a simple MAC

> person like me and I will raise a comment on D1.0!

> Bonus questions:

> - Is CCA-ED only required of the OFDM (11a) PHY? Or is it also

> required for PHYs which incorporate the OFDM PHY by reference (using

> the San Antonio wording)? What about 11n in GF mode, should this ever be done in a regdom where CCA-ED is mandatory?

> - Why is the CCA-ED threshold in a MIB variable

> (dot11OFDMEDThreshold)? The other energy detect mode's threshold is

> not in any variable (or is this dot11EDThreshold (but this is only for

> the DSSS

> PHY) or even dot11TIThreshold (deprecated and in the dot11PhyOFDMTable

> though its description seems generic enough)?)

> - What's the difference between "CCA-ED", "CCA-EnergyDetect", "CCA-Energy Detect", "CCA energy detect"

> and "CCA-EDBehavior"?

> Are the second and fourth just typos for the third, and the first just

> an abbreviation of the third; the fourth being the behavio(u)r associated with regdoms which require CCA-ED?

> Mark

> --

> Mark RISON, Systems Architect, Wi-Fi English/Esperanto/Français

> Samsung Cambridge Solution Centre Tel: +44 1223 434600

> Innovation Park, Cambridge CB4 0ZT Fax: +44 1223 TBCTBC

> ROYAUME UNI WWW: http://www.samsung.com/uk

IF YOU WISH to be Removed from this reflector, PLEASE DO NOT send your request to this CLOSED reflector. We use this valuable tool to communicate on the issues at hand.