Re: [ns] question about the shortest path

[Liang Qin <lqin@sce.carleton.ca>]

Hi there, 

Yesterday I ran the simulation using the scenario files generated by the
setdest.cc in the ns2 package. The problem seems gone. Only at most 2 such kind
of packets appeared in 65k total for DSDV, AODV and DSR protocols.

I don't know the reason, but that's the result I got.

Thank you for your help!

Liang Qin





On Thu, 24 Aug 2000, Satish Kumar wrote:
> Just a guess but it seems like the transmission range specified in setdest.cc 
> (by the RANGE variable, I think) may be different from the transmission range 
> used in the simulations (which is computed from the propagation model).
> 
> Satish
> 
> 
> > Ya Xu-
> > 
> > By "ad hoc routing" did you mean the 'actual' hop count 
> > achieved by the selected routing protocol when the simulation
> > is run, or the 'optimal' hop count predetermined by the
> > Floyd-Warshall algorithm in the setdest.cc file?
> > 
> > I thought the Floyd-Warshall Algorithm calculated the least 
> > cost paths for source-destination pairs using a connectivity matrix.  
> > If so, then the problem below is that the routing protocol 
> > in the ns run is finding a path that is shorter than the one
> > determined via the Floyd-Warshall algorithm with perfect
> > knowledge of the network connectivity.
> > 
> > thanks,
> > 
> > jeff