> > The nodes are actually quite far apart (I'm using high gain antennas and am > > actually getting around 50% signal strength). A is 1.2km from B and C is > > another 200m from B. B is at the top of a hill. A and C cannot reach each > > other directly (at least not with enough signal strength for a packet to > > pass through without error). A and C are also on different subnetworks. B > > routes packets between the two networks. > > I think you're stretching the capabilities of 802.11b here :-)
I'm not too sure this is actually stretching the capabilities. After all, cisco will provide you with all the hardware you need for link distances much greater than this for their 802.11 cards. > I'm not at all surprised you're getting problems if A and C cannot tell when > each other is transmitting - B is sometimes going to get a garbled mix of > signals from both, so unless you have any way of running B on two different > 802.11b channels to communicate with each of A & C (and I presume it's just a > single card, so you can't) this interference of signals is something you > can't get away from. > > Just out of interest, why are you running a setup like this in Ad-hoc mode > instead of Infrastructure mode ? I'm no technical expert on this (but I do > use 802.11b), however I can't help feeling that Infrastructure mode would be > better able to cope with two client machines which both talk to the access > point / base station, but can't hear each other... I was just hoping I could avoid using an access point as they are significantly more expensive than a wireless ethernet card. Maybe I should try and find out how access points avoid collisions. Russell
