[CUWiN-Dev] Cisco cards

[Quantum Scientific]

On Friday 04 March 2005 14:16, Jim Thompson wrote:
 > By the way, my understanding from one of the MIT Roofnet developers is 
> > that in regard to loss of throughput at each hop: "We're also 
> > experimented with 54mbit 802.11g radios and found they actually do 
> > worse than the 11b radios. "
> 
> I can think of several things that could cause that, most of them 
> fixable or avoidable.

This is because multipath has a much greater impact in OFDM.  I don't know 
what can be done about that, other than directional antennae, which implies a 
client/directional  server/omni node topology.

All this throughput loss per hop has been quite a concern for me.  My answer 
is two radios per node, because I will have quite a bit of linear hopping 
going on.

Carl Cook