Scalability of manets (was: Re: [OLSR-users] Question about OLSR

[aaron]

On Sat, 15 Jan 2005, Jens Nachtigall wrote:

hi!


What I meant was that - under ideal conditions (+ with something like LQ)
when each hop decides where to route each individual packet then the 
maximum that you can achieve (intentionally ignoring interference 
problems for now) would be given by the max flow algorithm.

However with interference: quality and thruput will diminish rapidly

We tried to use only directional links as much as 
possible to avoide interference but that also does not work:
   1) channels 5-9 are "spammed" by some other signal source
   2) if you try to jam 3 directionals onto one antenna pole,
      and you can't use channels 5-9 then you will get interference.
      The stronger signal wins.


I briefly googled for it:
http://mathworld.wolfram.com/MaximumFlowMinimumCutTheorem.html

http://www.lri.fr/~durr/MaxFlow/


I hope this mail was a bit clearer :)

ciao,
aaron.