Re: Wireless Simulation

[Padmaparna Haldar <haldar@ISI.EDU>]

Hi all,
      we are working to remove bugs in both dsdv and dsr
implementation. The code ported from cmu being considerable in size and
the changes made in core ns also being substantial, we were expecting bugs
and maybe should have declared this version of wireless code as
alpha-test only!
Hopefully we'll be able to set out patches or may even set out a special
snapshot for wireless users by next week. Your feedback on the bugs
[especially detailed reports of it] is extremely helpful to us and
goes a long way towards a faster stabilisation of the code.

Samir: we have removed the problem that you see as a difference between
the address representation in pkt hdrs.

Thanks,
--Padma.

On Wed, 31 Mar 1999, Verena Rose wrote:

> Hi all,
> 
> I have tried the example file ~tcl/ex/wireless.tcl in ns-allione-2.1b5
> (I'm a solaris user).
> With the given connection pattern ~tcl/mobility/scene/cbr-50-20-4-512
> and the scenario
> ~tcl/mobility/scene/scen-670x670-50-600-20, the output looks quite nice,
> the 50 nodes seem to move around. My problem is now, how to get any
> traffic into the simulation?
> I've looked into the connection pattern file, and it seems that traffic
> is generated when
> simulation has already stopped. I tried to extend the simulation time,
> but this was giving
> me "Segmentation fault", maybe because of too many movements causing too
> many lines in the output?
> I tried to change the times scheduled in the connection pattern file
> then, but there were
> only three packets of one connection visible in the trace file.
> How can I generate traffic of any kind for the mobility scenario?
> Anyway, is there some help about what lines in mobility trace files
> mean?
> 
> One more problem, when I tried to switch the logging of movement off,
> there was the
> following error:
> can't read "ns": no such variable
>       while executing
> "set ns_ $ns"
>     (procedure "dsdv-create-mobile-node" line 5)
>     invoked from within
> "dsdv-create-mobile-node $i"
>     (file "wireless.tcl" line 250)
> I've looked at line 250, but it seemed like a normal loop to me and I
> can't make anything out of the error. I didn't change anything there, I
> just set opt(lm) to "off".
> 
> Maybe someone can help me who has tried the example wireless.tcl and
> give me a hint how I can do a wireless simulation with the mobility
> code.
> 
> Thanks in advance,
> 
> Verena
> (rose@telematik.informatik.uni-karlsruhe.de)
> 
> 

------------------------------------------------
Be true to your work, your word, and your friend.
--Thoreau

Padmaparna Haldar
ISI/USC.
310.822.1511 #352