Last Post: November 29:
Take a peak at this FCC document and look at the FCC and other docs it references Eg. CFR's, etc.
Google IXIA and DFS.
You'll have to do the legwork.
Thanks Howard! That's an excellent example of testing arbitrary waveforms and RF test setup design to simulate a real world environment !
Also take a look at the recent post by mali27 on "DFS and Airport"..
Now that DFS is being enforced by ETSI and FCC I've seen plenty of false positives. Earlier most AP vendors didn't even bother implementing DFS or it was turned off by default. Now it is being tested as a part of the certification. If the AP doesn't change channel it won't be allowed on the market. There is no penalty for false positives however, so the pressure is biased.
At least Ubiquiti is working on improvements, but that is still work in progress. It appears to be a difficult problem to solve. I just had to look into this for a client. They have quite frequent channel changes but all indoor APs on the same channel don't change channel at the same time, so I believe they are false positives. Of course it could be a chopper flying by brushing just a corner of the building, but they occur in the middle of the floor as well.
The best (short) resource I have found is
which includes the different pulse patterns and other requirements.
That IS a great resource. It puts so much technical detail in one place.
One section describes something that really bothered me when it was announced :
" Devices are prohibited from including configuration controls (e.g. country code settings or other options to modify DFS functions) to change the frequency of operations to any frequency other than those specified on the grant of certification for US operation."
Depending on how this was implemented internally, it could really cost manufacturers a ton of money to implement, and could reduce worldwide market possibilities greatly. The original Country Code controls were really trounced by this pronouncement.
I would like to get a copy of the ETSI document they mentioned, TR 102 651, but I don't have access to it and the only place I found it wanted $50 for it. I'm not that interested in it.
UPDATE: ETSI is publicly available if you know where to look. I found this document at: http://www.etsi.org/deliver/etsi_tr/102600_102699/102651/01.01.01_60/tr_102651v010101p.pdf
Devices are prohibited from including configuration controls to change the frequency of operations to any frequency other than those specified on the grant of certification for US operation.
I read that as if the vendors change configuration controls (add new countries or new options affecting DFS) after the certification then they need to recertify the device. I don't know if they need to recertify after each firmware change while they are trying to weed out the false positives. That is not a configuration control, though.
You can still set all devices to South-American countries (Paraguay, Uruguay, Bolivia IIRC) that don't have any restrictions on WiFi frequencies. I guess that text just prohibits adding a new country like US-noDFS to the list or a switch to turn DFS off.
(IANAL and all other disclaimers apply)
What has happened is that some companies end up making country specific products that don't allow country codes to be changed or ignored ! Hard coded channels control it all - no 802.11h.
If you are selling to over 100 countries, that can make doing business a real pain.
Regulatory agencies and customs officials hold the manufactures responsible for meeting their standards. They sometimes tend not to care if some unscrupulous 3rd party was involved with a shipment. If your devices are held at a border by customs, it could be held for weeks, or months and cost big bucks to get the shipment out of their clutches. Big companies would rather not takes such risks.
Some countries are worse than others, and different ports of entry may have different opinions of what their countries laws really are. - Seems impossible, but it's true.