Hi!

For some reason I never saw the original post... maybe the spam filter 
ate it.

Rest of comments inline.


David Fuste schrieb:
> First of all, thank you very much to all...I thought the post was lost 
> :) Martina, your information has been very useful.
> And second, thanks Pedro, I will look for this code in the cvs that 
> partially supports infrastructure WLAN simulation.

Interesting. I was not aware of this code but I am not doing WLAN 
simulations myself anyway. Can you post your experiences when you use 
it? Might be helpful for a lot of other people, too.

I've always wondered why there wasn't an obvious way to do 
infrastructure simulations but apparently people were only interested in 
simulating the ad-hoc stuff.

I do think that infrastructure WLAN is an interesting scenario, though, 
esp. if you could couple it with, say, 3G and do simulations of 2 cells 
and handovers between them... <daydreaming> ;-)

[snipped for brevity]

>> I 've also heard abt NOAH but I can find any documentation on ow to 
>> use it.
>> I mean, the way it is explain; says a node can send  a message to a 
>> BS(Base
>> Station) but details as to ow to send a msg to anoda node thru the BS 
>> isn't
>> included (I mean, it's rare to keep sendin msges to a BS unless one 
>> intends
>> to send to anoda mobile node).

Hmmm. I've only ever made simulations (actually not myself, a student of 
mine did) with WLAN + NOAH where traffic comes from wired host nodes, 
goes through the AP and then to the mobile nodes. Iow. a typical 
wired-cum-wireless scenario. I have never actually tried to have traffic 
from one wireless node to the other.

Basically, NOAH is a routing module that allows you to manually set 
routes. It does _not_ change the MAC layer in any way. So, I believe 
that if you set up NOAH and set up the routing table accordingly it 
/should/ work if you simply set the traffic agent on one mobile node and 
the sink on the other. The traffic should then be forced by the routing 
to first go to the BS and then go to the other mobile. I can't remember 
if the MAC layer is/should be significantly different for infrastructure 
mode as compared to ad-hoc mode but if you can live with the MAC as-is 
you should be able to get what you want with NOAH.

>>
>> Also the issue of MIPv6 is still hanging, the mobiwan extension is 
>> claimed
>> to be workin on NS-2.1b6 and above but I am not sure if it does work with
>> NS-2.29. Pls, can u confirm if it does work wit NS-2.29?

Based on my experiences with ns I'd say no. There have been many changes 
  between 2.1b6 and 2.29, a lot of them to support gcc 4 so that alone 
would probably break the patch many times over. If you are really 
dependent on a certain patch it's probably faster to get an older 
version of Linux and use that old version of ns. I myself have all my 
machines installed with a previous version of Linux to support ns-2.26 
which needs gcc < 4 because of a patch that I use...

OTOH, there also have been lots of bugfixes in the meantime, so it might 
be worth the effort to port a patch to 2.29. I mean, 2.1b6 is like from 
the stone-age... ;-)

HTH,

Martina

Reply via email to