Forum

also the sh contr dot11 0 and 1 output.

I'll PM you Frans.

Sometimes the military and Government agencies will conduct tests in the ISM and UNII type bands without telling anyone. It doesn't happen that often, but has happened in the past. Usually, they will just use one band ( most often the 5 GHz band ). You are seeing it on both the 2.4 GHz and 5 GHz bands. That would be extremely unusual for mil/gov testing to take place in both bands.

This could well be due to solar activity. One thing that would be interesting would be to check at nighttime when the part of the earth where you are is facing away from the sun. Unfortunately, that can't be used as a major discriminator of solar interference/terrestrial interference, as many test systems are switched off at night. Some do run for a few days or weeks 24 hours a day.

Dave

Frans

Tried to PM you, but don't see any contact details ( mail won't go through ). You might have to go into your CWNP account. I know there were some problems with PMs a while back.

Dave

I PMd you. Today resumed testing. Spectrum analyzer has been running all day on different channels. I do see peaks today on the UNII-1 channels, so I moved away to UNII-2e. I also saw big activity in the 2.4 GHz once, all over the ISM with duty cycles up to 100%. During testing I have the spectrum analyzer running at the same time.
I have been testing with no encryption as well today. .11a/g works fairly good, not in all tests. HT doesn't show up at all in the 2.4 GHz. On channel 112,-1 HT worked ok for a while, channel 100,1 gave me up to 118 seconds between ICMP replies (nothing much going on in the spectrum analyzer). ping response times in the dosbox however show between 3ms and 90 ms (so a lot of variation I'd say).
This has been going on for all days, the simpler de modulation, the best chance I see the frames, the more complex, the more they disappear. Somtimes a frame is shown in Airmagnet, and sometimes in wireshark, sometimes both if I get lucky ;-)

Update: Testing between 7 pm and 8 pm today gives much better results. That was not the case yesterday.

Fran

Got your PM. You don't happen to have any "snapshots" of the interference from the analyzer, do you ?

Dave

sure do. I can send them.

Now for the analysis, I was thinking of the following:

To determine the number Acks with DUT as RA:
wlan.ra == 00:22:fa:3e:91:b2 and wlan.fc.type_subtype==0x1d
There should be twice as much dataframes in the traces resulting from the Acks minus the Null frames:
wlan.sa == 00:22:fa:3e:91:b2 and wlan.fc.type_subtype==0x2c
And then compare to the actual visible dataframes
wlan.addr == 00:22:fa:3e:91:b2 and wlan.fc.type_subtype==0x28

This should give a good indication with maybe a deviation of a few.

The plots are very interesting. If you have time for a Skype chat, let me know, as there is quite a lot to discuss.

There is another possible reason:

http://www.youtube.com/watch?v=zVOAEZTEVKA

http://www.youtube.com/watch?v=-ha771uZK-k&feature=related

http://www.nato.int/docu/awacs/awacs-e.pdf

For a number of reasons ( many of which are discussed below, frequency bands do not have to match ). The London Olympics are coming up and there will be a lot of airborne activity. The aircraft is capable of multiple missions and contains a large amount of classified signals equipment. NATO has a number of these aircraft, with Dutch crews being involved. The aircraft can ?loiter? or remain flying in small circles over one area for long periods of time. With airborne refueling, and other aircraft working together, almost constant patrolling can take place. Unfortunately, if that is the reason, the chances of getting information from NATO is less than zero.

http://www.biroengineering.com/technical_bulletin/tb_2002_01.html

http://www.biroengineering.com/technical_bulletin/tb_2002_04.html

http://www.comsearch.com/newsletter/archiveWP/WirelessPulseJan02.html

http://www.drf.com/news/theory-simulcasts-zapped-awacs

http://www.eng-tips.com/viewthread.cfm?qid=15231

Dave

Thank you Dave, interesting. the AWACS radar could bleed on different frequencies, however should not cause long term problems on a specific frequency.
It could be that the radar dots have a relationship to the laptop it was attached to. The antenna is attached to the laptop and the WLAN antenna cables within the screen are 1 inch away. I have turned it off and the spectrum becomes completely quiet. Problems however are still there.

I'm available for discussing the interpretation of the plots today between 6 and 10 PM (GMT+1 here) and tomorrow whole day.

Frans

Please let me know your Skype username by PM. I sent mine to you. Will leave Skype on today and tomorrow.

I can't go into details, but many of the Airborne platforms such as AWACS are capable of other functions besides radar transmission and reception. The 10s radar antenna turning time does not apply in those modes.

Have you ( or the other chap at the location 80 miles away ) used a directional antenna well away from the hospital ( horizontal scan ) to see if the problem is terrestrial based. In other words, a check to see if you get the same problems well away from any possible hospital equipment.

Was the other chap at a hospital or some other facility ?

Is there any equipment with DFS, so we could have a look at logs and see if there were any channel changes etc

Dave