On Mon, 3 Jan 2005 14:55:07 -0500, "Jonathan"
<(E-Mail Removed)> wrote:
>I'm sorry, this is not correct. There is absolutely no spectrum overlap
>between 802.11b/g channels 1, 6 and 11. Check it out for yourself in any
>good WiFi book. This is where the "5 channels" rule-of-thumb originated.
Actually there is, because the spectra extends well beyond the alleged
occupied bandwidth. Maybe an explanation will help.
802.11b/g direct sequence channels are 5 MHz wide[1]. 802.11b/g
direct sequence spread spectrum (DSSS) modulation looks like a
sin(x)/x curve, with a big major "lump" in the middle, a drop to zilch
on either side of the lump, and then smaller lumps going of into
infinity. See tutorial at:
http://www.sss-mag.com/ss.html#tutorial
The photo of the DS spectra is what I'm muttering about.
The width of the "lump" in the middle is about 22MHz wide. The bulk
of the RF is concentrated in this 22MHz or about 4+ channels wide. 5
channels per signal is a good average.
However, much of the RF on the slope of the "lump" is relatively low
level when compared to the peak power. If you have sufficient
physical seperation to reduce the signal even more, it is possible to
run a system on perhaps every 3 or 4 channels seperation instead of 5.
However, if the radios are in the same "air space" and can hear each
other, every 5 channels is considered safe.
[1] Frequency hopping spread spectrum channels a 1MHz wide.
--
Jeff Liebermann
(E-Mail Removed)
150 Felker St #D
http://www.LearnByDestroying.com
Santa Cruz CA 95060 AE6KS 831-336-2558