When we have a "G" network, and a "B" client is detected, at what point is protection triggered ?
I have been reading conflicting info, two white papers with different views...
There are two options as I see it - when the "B" station is in discovery, the probe request will kick the AP into protection.... OR...
The AP responds with a probe response and but doesn't trigger protection until the "B" station associates.
I need to get some lab time to check, but until then, what's the consensus ?
(I have always been of the understanding that merely detecting a "B" client invokes protection)
Any views ?
An 11g AP will go into protection because of an Non ERP Present (11b device, which can include an associated 11b STA or an 11b AP). To indicate to the BSS that it is in protection, it will set two bits in the beacon. (This is two of the three ways an 11g STA will go into protection.)
Non ERP Present = 1
Use Protection = 1
Now for the third, and most complicated. If an 11g AP hears an 11g AP with the above bits set, it will also go into protection. However, since it doesn't actually see an 11b device it will set the following bits:
Non ERP Present = 0 (It doesn't detect an 11b device)
Use Protection = 1
Now you ask, what happens if an 11g AP sees an 11g AP with the above bits? Nothing. It won't go into protection.
This is a bit hard to explain in a post so let me know if I can clarify.
Yeah, I know all that, however at what point does it set those bits ?
This from an eminent white paper:
"As soon as a NonERP STA associates to an ERP AP, the AP will signal for protection in the ERP information element of its beacon...."
This from a well known book on 802.11:
"Protection is not activated by associations, however, but by the detection of 802.11b transmissions. All it takes is one 802.11b station in the area."
There are plenty of other examples saying it only kicks in AFTER association.
So there are two schools of thought.
Gene, I think you are on the non-association side ?
Sorry I didn't read your original post thoroughly.
Interestingly enough, we just spent an hour or so working on protection stuff last night. My guys in class wanted to see a few things:
This was with Motorola Equipment.
1. If an AP on channel 1 goes into protection, will all other AP's (on other channels) on the controller go into protection? The answer was no, which was expected but they had heard something else.
2. Would an AP in "11g" only mode and only having OFDM data rates enabled go into protection because of another 11g AP in protection? The answer is yes, although this is unique to Motorola equipment.
With you asking that question, we actually answered it last night and I didn't even realize it until this morning.
We had two AP's with the same SSID, one on channel 1 and the other on channel 6. We connected an 11b device to the AP on channel 1 and that AP went into protection just like expected.
I did witness the 11b device proactively probing, just like it is supposed to, however, that action did not send the AP on channel 6 into protection.
I do think that different vendors may take different action, but I am happy that the Moto gear didn't go into protection because of the probes from an 11b device. In 90% of all 11g environments, protection is a hindrance, not an asset.
Um... can vendors do that ? Isn't it part of the standard ?
So, this points to the fact that protection will only kick in when the "b" station associates, not when it probes.
Yep! Vendors can do that. If I remember correctly, "11g only" mode isn't standardized and many vendors do it.
From what I have seen, the 11b STA must associated. However, keep in mind that an 11b AP will also invoke protection.
I get a lot of questions when you see both bits set to 1 but no 11b device is associated. That is because of the 11b AP can invoke protection.
Yep, so it has to be associated.
Here is the (incorrect) info from Matthew Gast, author of 802.11 Wireless Networks: The Definitive Guide (O'Reilly press)
Top 10 802.11 Myths:
Myth No 5:
802.11g protection forces all data to be transmitted at 802.11b rates.
Protection is the protocol feature the gives 802.11g compatibility with older 802.11b equipment. It does slow down 802.11g stations, but not by forcing them to use slower 802.11b data rates. When an access point activates protection, it "wraps" faster 802.11g transmissions with a slower, backwards-compatible frame. The slowdown comes from the backwards-compatibility wrapper, not a reduction in the data rate of the frame.
Related to this myth is the belief that only allowing 802.11g stations will improve the data rate of an 802.11g network by preventing slower stations from associating and triggering protection. Protection is not activated by associations, however, but by the detection of 802.11b transmissions. All it takes is one 802.11b station in the area. Given the prevalence of 802.11b in the installed base, protection will be active on most networks for the foreseeable future.
So, he's wrong then.
(By the way, I agree with you that it IS after association, after reading up on whitepapers).
I'm confused here.
If the one AP in a BSS goes into Protection Mode, does the entire ESS go into protection mode even though other BSSs don't have a 11b device associated to them?
If yes, then what purpose does this serve ?
Because according to GTHill's post that was on the front page
"Other 11g APs in the area will also use protection, even if they don?¡é?€??t have an 11b device in the area."
And from his lab result on the Moto AP, the second AP on the different channel didn't go into protection mode when a 11b device associated to the first access point.
So I'm definitely missing something here :( .