• I have seen several military/governmental systems which have caused problems with a variety of communications equipment at various bands.

    I worked at a satellite communications facility in a desert. We were hit each day with broad band interference. It was one of my first jobs. I looked on the spectrum analyzer and said that it was a military radar and jammer. I was just a junior engineer at the time and nobody paid any notice of what I said. They had a good laugh and let me know that we were in a desert. It was a serious problem, and the customers were going through the roof. TV programs were being interrupted. They flew experts out from the UK to no avail. One day, one of the night watchmen came in saying that he saw a man coming out of a hole in the ground through a hatch. He went up to it and was chased away. Way before the first gulf war, a country that has been in the news recently was getting ready to invade the location where I was. A friendly power had come in the middle of the night over a few weeks and installed a bunch of radar and jamming equipment in dug out areas, complete with diesel generators etc. They'd worked at night. Prior to that, the night watchmen had kept saying they'd heard vehicles and occasional helicopters. Nobody paid them any heed. I went out to the site and it was beautifully camouflaged. Spay painted fiberglass.

    The invasion never took place, as locals who were involved with the other country were caught in time at a meeting place.

    A call to the appropriate embassy and the interference stopped.



  • Frans

    Please send me the packet traces to my e-mail. Haven't received them yet.



  • The problem is that I cannot see a direct relation in the spectrum analyzer and missing frames in the traces.
    The hospital is using nonDFS UNII-1, to forecome any radar-caused channel switches. I can confirm the dots are related to the testcomputer using that wlan antenna as well. I have traces from the hospital where I don't miss a single frame. I have a 3502 over here and no reports whatsoever about radar.
    Reports come from 3 geographical different locations in the netherlands (east, south and west).
    Fridaymorning it was about to look good and friday afternoon it was very bad again. Right now it is going ok and I see most of the frames.

  • the other thing is that if there would be radar, all traffic should be deferred, but looking at the Acks I see that the traffic continues (also from the pings and transfers I have been doing). I have to analyze all the traces of the last few days, but one thing is for sure:
    1. There has been some increase in dataretries
    2. There has been a huge impact on the protocol analyzers. Especially with decoding the more complex modulations.

    If there were radar bursts, the more complex modulations would probably suffer less because of shorter transmission times.

  • Sorry, a bit to optimistic. Only 5 GHz HT shows up. 2.4 HT not (a handfull)

  • As to why your analyzers may be more affected than some of your regular wlan equipment, is there any chance it is due to the analyzer gear having better sensitivity ?

    I have seen a lot of g/n radios numbers that vary by as much as 20 dB.

    Have you tried the analyzer in different PC's?

  • Hi Frans

    Nice chatting with you last night. You live in a beautiful area. I've always liked Holland and Norway. My unit used to do a drop at Ginkel Heath each year to celebrate Operation Market Garden. They used some WW2 Dakotas ( C-47s ) and modern C-130s. I missed it three years in a row from injuries from other jumps. Finally managed to visit Arnhem after another airborne exercise in Germany. The grave of each paratrooper is tended to by a Dutch child. The graves are immaculate. This duty is onpassed to their children. The gratefullness of the Dutch people was very touching.

    There is definitely something strange going on. The fact that you and your colleagues have used different adaptors/laptops/analyzer programs at different locations and you got the same results in an electrically quiet area seems to suggest that there is some form of broadband interference going on.

    I'll give Fluke a call in a bit and have a chat about about the patterns.

    Will leave Skype on all day if you want to chat later.


  • Just waiting for a callback from Fluke.



  • Would think so. Have 3 laptops, 8xAirpcapNX, 2xProxim, 1xC1060 and PRISM Duette to analyze with. Use them with Wireshark, AM analyzer and Ethereal (prism). AP is about 7 meters away, analyzer adapters either in laptop or with usb cable up to 1 meter away at different positions. I have 1GB material now from the last days. I will analyse this the coming days to see if I didnt make an error myself (probably not).

    Something weird to think about (doesnt always happen but observed it by accident):
    AP on channel 1 20MHz.
    Airpcap 1 set at channel 1 20 MHz
    Airpcap 2 set at channel 1,-1 40MHz, this one I see a few more frames.
    Start 2 different sessions of Wireshark. I used to miss alot of frames in 11g as well the past days, but lately 11n is the last hurdle.
    I move with my hand over the laptop cover where the antenna cables are integrated and suddenly more frames pop up (no DRS because it also happens when there is no rate change).
    There might be something with polarization and/or synchronization.
    5GHz 11n is in de mean time showing much better results

Page 3 of 4