snipped-for-privacy@panix.com (Al Dykes) hath wroth:
Naw, I goofed. I was running two threads involving spectrum analyzer issues and managed to get them muddled. If you're curious, see:
for the actual question under "Quiz Time".
It will wander around about 6dB without much effort and 10dB if you start to move things around. Ever notice those nifty looking Netstumbler signal strength and SNR graphs? Ever notice that they're NOT flat top lines, but are rather ragged on the top? Those are the variations I'm mumbling about. For example, look at Netstumbler screen at:
Notice the variations over 8 minutes (total) run. If I ignore the total dropouts, my guess from the two parts of the graph are 12dB variation on the left and about 10dB on the right.
Try it. Place a laptop somewhere that it's not going to move, and probe an access point (preferably your own) that also isn't going to move. Start it running and then go away for a while (to avoid having your body affect the readings). As long as you don't cheat and use an ultra strong saturating signal, you'll see variations similar to the above URL. The numbers are real and you'll also see them with a spectrum analyzer, on an antenna range, and in any highly reflective environment. You might not see it in an RF anechoic chamber, but I currently don't have access to one of those to try it.
Hmmmm.... I just fired up Netstumbler and checked the neighbors access point (mine is too close and too strong). I'm seeing over 10dB variations over about 3 minutes if I ignore the dropouts.