Last Post: November 29:
With the push towards much greater use of the 5GHz band due to the advantages of channel bonding in 11n and the forthcoming 11ac standard, which is exclusively 5GHz, I've been considering the challenges of using the 5GHz band more recently (in addition to the obvious advantages!).802.11h describes the DFS mechanism for radar detection and avoidance that must be employed by various channels in the 5GHz band (which vary from country to country).
My question is this: how often do folks out there actually see the radar avoidance mechanisms kick-in and cause an AP to change channel?
I know this is a bit of a loaded question, as there may some folks who are close to radar installations who suffer from it regularly. But for the general town or city-based WiFi network, maybe in office buildings, a college campus, a hospital etc. that isn't next to an obvious source of radar, does anyone see it that often?
I personally have only seen it on one occasion over the past 5 years or so when some Cisco APs reported a radar blast detected whilst attending a training course in a building near a local airport.
Is it a regular occurrence, or is it something you are pretty unlucky to hit generally.
I know that it is difficult (and dangerous) to try to generalize about these things, but I am interested to hear other folks experiences of this.
I don't have any first hand DFS field experience, but I have heard that indoor 802.11n networks in the IIe band will not usually see Radar pulses with properly spaced and configured AP's. i.e. no high-gain antennas & all AP's far from the periphery, etc.
Very different outdoors though - if you're asking for it in other than a rural environment.
As an aside, I know it can be difficult for Wi-Fi manufacturers to properly detect Radar signals during testing. The signals must really look like Radar and be very short. Veriwave makes modules for their gear that wiil let you test for it, but an entire system is lots of $$$.
I suggest you search for 802.11n in the Vocera Technical Support Knowledge Base. Sorry I don't have the link. Good luck.
Here's some info on what you need to comply with on 5600-5650 MHz band in addition to the other DFS channels. Vendors I know sit on a council and were are in talks with the FCC to allow operation in the excluded zones, but they seem to have a pretty archaic view on how to regulate it. To be safe, stick to the 35km exclusion.
ME M O R A N D U M
From: Julius Knapp, Chief, FCC Office of Engineering and Technology
P. Michele Ellison, Chief, FCC Enforcement Bureau
To: Manufacturers and Operators of Unlicensed 5 GHz Outdoor Network Equipment
Date: July 27, 2010
Subject: Elimination of interference to Terminal Doppler Weather Radar (TDWR)
The Federal Communications Commission (FCC), the National Telecommunications and
Information Administration (NTIA) and the Federal Aviation Administration (FAA) have
been investigating interference caused to Terminal Doppler Weather Radar (TDWR)
systems operating in the 5600-5650 MHz band. TDWRs are used to detect wind shear
and other weather conditions near airports. The interference manifests itself as a strobe
line or lines on the radar display. While the radar continues to be usable, such
interference is unacceptable and must be eliminated. More importantly, if the
interference were to become severe, there may be a potential for missed alerts or false
We have found that the interference at each location has generally been caused by a few
fixed wireless transmitters used by wireless internet service providers (WISPs) and
operating outdoors in the vicinity of airports at high elevations that are line-of-sight to the
TDWR installations (5 GHz outdoor network equipment). In most instances, the
interference is caused by operations in the same frequency band as TDWRs, but there are
some instances where the interference is caused by adjacent band emissions.
The 5 GHz outdoor network equipment is required to be certified under the FCC rules
governing Unlicensed National Information Infrastructure (UNII) devices. In some
instances, the interference was caused by equipment that was not certified or otherwise
was not compliant with FCC rules. The FCC has taken appropriate enforcement action in
those cases. We remind operators and manufacturers of UNII devices that any use or
marketing of equipment that has not been certificated as required under the FCC rules or
that has been modified such that it no longer complies with the certification requirements
will result in FCC enforcement action.
In other instances, equipment that met the FCC’s certification standards nonetheless
caused interference, due to a variety of factors such as: the configuration of the
transmitter, its height and azimuth relative to the TDWR, and the device’s failure to
detect and avoid the radar signal. In such cases, the FCC’s rules still require the
elimination of the interference and the FCC has taken appropriate action in those cases.
The FCC, NTIA, and the FAA have been working with manufacturers of UNII devices
and the Wireless Internet Service Provider Association (WISPA) to fully understand the
causes of the interference problem and to develop effective and targeted remedies.
Through our discussions, the various parties have agreed upon a number of immediate
steps:The FAA has provided information on the locations of each of the TDWRs (see attached).
The Wireless Internet Service Providers Association (WISPA) has voluntarily agreed
to disseminate the location of the TDWRS to WISPs.1 Moreover, WISPA has agreed to encourage operators that install devices within 35 km or the line-of-sight of the TDWRs, to operate at least 30 MHz away from the TDWR operation frequencies.
WISPA has also agreed to voluntarily provide a database where WISPs can register
the locations of the outdoor transmitters that they use. This database will be helpful
for identifying sources of interference if it should occur. 2The manufacturers of UNII devices involved in our discussions have also agreed to
conduct outreach to their customers to alert them to the steps they must take to ensure
that they do not cause interference to TDWRs. We urge all manufacturers of UNII
devices to do the same. These steps include avoiding operation in the TDWR band
and operating at least 30 MHz away from the TDWR operation frequencies when
installing devices within 35 km or the line-of-sight of the TDWR sites.
The FCC, NTIA and FAA are continuing to vigorously investigate and eliminate cases of
interference to TDWRs. It is our expectation that with your cooperation all such
interference can be eliminated.
TDWR Location Information*
STATE CITY LONGITUDE LATITUDE FREQUENCY
TERRAIN ELEVATION (MSL) [ft] ANTENNA HEIGHT ABOVE TERRAIN [ft]
AZ PHOENIX W 112 09 46 N 33 25 14 5610 MHz 1024 64
CO DENVER W 104 31 35 N 39 43 39 5615 MHz 5643 64
FL FT LAUDERDALE W 080 20 39 N 26 08 36 5645 MHz 7 113
FL MIAMI W 080 29 28 N 25 45 27 5605 MHz 10 113
FL ORLANDO W 081 19 33 N 28 20 37 5640 MHz 72 97
FL TAMPA W 082 31 04 N 27 51 35 5620 MHz 14 80
FL WEST PALM BEACH W 080 16 23 N 26 41 17 5615 MHz 20 113
GA ATLANTA W 084 15 44 N 33 38 48 5615 MHz 962 113
IL MCCOOK W 087 51 31 N 41 47 50 5615 MHz 646 97
IL CRESTWOOD W 087 43 47 N 41 39 05 5645 MHz 663 113
IN INDIANAPOLIS W 086 26 08 N 39 38 14 5605 MHz 751 97
KS WICHITA W 097 26 13 N 37 30 26 5603 MHz 1270 80
KY COVINGTON CINCINNATI W 084 34 48 N 38 53 53 5610 MHz 942 97
KY LOUISVILLE W 085 36 38 N 38 02 45 5646 MHz 617 113
LA NEW ORLEANS W 090 24 11 N 30 01 18 5645 MHz 2 97
MA BOSTON W 070 56 01 N 42 09 30 5610 MHz 151 113
MD BRANDYWINE W 076 50 42 N 38 41 43 5635 MHz 233 113
MD BENFIELD W 076 37 48 N 39 05 23 5645 MHz 184 113
MD CLINTON W 076 57 43 N 38 45 32 5615 MHz 249 97
MI DETROIT W 083 30 54 N 42 06 40 5615 MHz 656 113
MN MINNEAPOLIS W 092 55 58 N 44 52 17 5610 MHz 1040 80
MO KANSAS CITY W 094 44 31 N 39 29 55 5605 MHz 1040 64
MO SAINT LOUIS W 090 29 21 N 38 48 20 5610 MHz 551 97
MS DESOTO COUNTY W 089 59 33 N 34 53 45 5610 MHz 371 113
NC CHARLOTTE W 080 53 06 N 35 21 39 5608 MHz 807 113
NC RALEIGH DURHAM W 078 41 50 N 36 00 07 5647 MHz 400 113
NJ WOODBRIDGE W 074 16 13 N 40 35 37 5620 MHz 19 113
NJ PENNSAUKEN W 075 04 12 N 39 56 57 5610 MHz 39 113
NV LAS VEGAS W 115 00 26 N 36 08 37 5645 MHz 1995 64
NY FLOYD BENNETT FIELD W 073 52 49 N 40 35 20 5647 MHz 8 97
OH DAYTON W 084 07 23 N 40 01 19 5640 MHz 922 97
OH CLEVELAND W 082 00 28 N 41 17 23 5645 MHz 817 113
OH COLUMBUS W 082 42 55 N 40 00 20 5605 MHz 1037 113
OK AERO. CTR TDWR #1 W 097 37 31 N 35 24 19 5610 MHz 1285 80
OK AERO. CTR TDWR #2 W 097 37 43 N 35 23 34 5620 MHz 1293 97
OK TULSA W 095 49 34 N 36 04 14 5605 MHz 712 113
OK OKLAHOMA CITY W 097 30 36 N 35 16 34 5603 MHz 1195 64
PA HANOVER W 080 29 10 N 40 30 05 5615 MHz 1266 113
PR SAN JUAN W 066 10 46 N 18 28 26 5610 MHz 59 113
TN NASHVILLE W 086 39 42 N 35 58 47 5605 MHz 722 97
TX HOUSTON INTERCONTL W 095 34 01 N 30 03 54 5605 MHz 154 97
TX PEARLAND W 095 14 30 N 29 30 59 5645 MHz 36 80
TX DALLAS LOVE FIELD W 096 58 06 N 32 55 33 5608 MHz 541 80
TX LEWISVILLE DFW W 096 55 05 N 33 03 53 5640 MHz 554 31
UT SALT LAKE CITY W 111 55 47 N 40 58 02 5610 MHz 4219 80
VA LEESBURG W 077 31 46 N 39 05 02 5605 MHz 361 113
WI MILWAUKEE W 088 02 47 N 42 49 10 5603 MHz 820 113
Latitude and Longitude are specified in NAD 83
* Last updated July 12, 2010
I live in Houston Texas and have hospitals positioned around Houston. I've seen regular DFS channel changes. We have recently moved away from UNII2 - UNII2E to lessen the impact of channel interruption.
Thanks for the information - very informative.
I'm surprised the list of locations was so short.
Hi Every one,
I am new to this 802.11 standards. I am presently going through DFS section. Many regulatory domains require "In service monitoring". May I know what does this exactly mean? Does it asks for detection of radar signal during normal reception of WLAN packets. Is it possible to detect presence of radar signal from the received combination of Radar and WLAN signals with DFS detection threshold of -64dBm (as per regulatory domains).
Or does it says that WLAN system needs to schedule radar detection during In service monitoring such that AP is in ideal state and does not receive or transmit WLAN packets, so that if any presence of Radar above the threshold can be detected.
Which of the above perceptions are true as per In service monitoring is considered?? Please help me to have a clear idea with your valuable comments... If available please quote some references.
As I have been struggling hard, to sort it out what does that actually mean, please help me to resolve it.
Thank you very much in advance...
I can tell you that even Compliance departments get confused over these regulations.
To be more precise, my doubt is that - As per WLAN standards minimum received signal detection threshold is -82dBm and as transmitter power is 20dBm(100 mW), the signal strength at the receiver can be -30dBm. That is, At receiver WLAN signal strength can be between -82dBm to -30dBm.
And as per regulatory domain requirements DFS threshold is -62dBm.i.e., radar signals above this power level must be detected.
Now assume a -40dBm WLAN signal and -50dBm Radar signal arrives at the receiver. In such a situation as Radar power is above the DFS detection threshold, it need to be detected. But, when it arrives along with WLAN signal of -40dBm power, Is it possible to distinguish the both and find the existence of radar signal component in the received samples??
I am very grateful to you if you could enlighten me about the possibility of above situation..
Anticipating help from you guys. Thank you in advance...
Good question. I'd like to know the answer too.
These days, I assume it will be able to be detected. The Radar signal will be a pulses, whereas the WLAN signal will be modulated data. However the pulses will be very short in relation to the data signals and that will be a problem. I was speaking to Veriwave several years ago when they were planning of adding a simulated Radar pulse to their wireless test gear and they said that they had found it to be a difficult problem.
Just to let you know, even though you quote the standard's limits, more realistic discernable WI-Fi signal limits, in newer products, will be between -20 and -90 dBm - both a lot stronger, and weaker than the original standard requires.
I consider -40 dBm to be a heck of a strong signal. You're essentially within a few feet of an AP at that point.
Thank you very much Howard for expressing your ideas on this issue. I would like to hear from you what can be the best way to find answer for this question. As you have mentioned it may be possible these days, it would be of great help if you could quote some references to go through. And Thank you for enlightening with much more realistic power levels.
Anticipating few more suggestions or recommendations on this issue.