da...@lang.hm wrote: > On Tue, 12 Jan 2010, Richard Chycoski wrote: > >> da...@lang.hm wrote: >>> On Tue, 12 Jan 2010, Richard Chycoski wrote: >>> >>>> I would also configure all of the APs for 'b' only (no 'g') to get >>>> the maximum use of your channel space. Mixing AP types as you have >>>> suggested may cause you more headaches, but then since $WORK builds >>>> our own, it's easy for us to use all the same AP everywhere - but >>>> it sure does make WiFi roaming between the floors (and buildings) >>>> less troublesome. >>> >>> b and g use the same channels, just at different speeds. I would >>> expect that everyone would be using g nowdays, and the more people >>> who use g the shorter the transmissions, and the less cluttered the >>> airwaves >> g uses groups of b channels. If you restrict the network to b, you >> get more channels to work with, and providing a cell structure is >> more practical. Besides, if any b devices come onto the net, it >> becomes b anyway. Any g device can operate as b, and I think that >> throughput is going to be the least of your worries - channel overlap >> and contention are likely to render g pretty well useless. > > no, g and b have the same RF footprint, in both cases each channel is > wide enought that it overlaps several adjacent channels. To avoid this > interferance the only channels that you can use are 1, 6, 11. It > doesn't matter if you are using b or g. My bad - somehow I had developed an incorrect understanding of the b/g channel arrangements. I sit corrected!
You *can* use overlapping channels if they are far enough apart, it's still better than using the same channels too close together. I'll see if I can get the bandplan that we use at work, because there are far too many APs on the floor for them to have been restricted to three channels. This paper <https://upcommons.upc.edu/e-prints/bitstream/2117/1234/1/CrownCom07_CReady.pdf> has a discussion on the use of overlapping channels that you might find useful. Spread-spectrum does help. Using overlapping channels will reduce throughput, but same-channel interference is much worse (throughput is very poor). - Richard _______________________________________________ Discuss mailing list Discuss@lopsa.org http://lopsa.org/cgi-bin/mailman/listinfo/discuss This list provided by the League of Professional System Administrators http://lopsa.org/
