Channels Wifi

[lluis....@gmail.com]

Hi,

I have a new question for Network Simulation. I'm work with NS 3.2
version but i don't know how can I change the channel tranmission wifi
in diferents APs... I read in another post, that this option is only
posible in NS 3.3, it's true?
And how can I change it?

Thanks for all,

Lluis

[Mathieu Lacage]

On Tue, 2009-03-31 at 06:25 -0700, lluis....@gmail.com wrote:

> I have a new question for Network Simulation. I'm work with NS 3.2
> version but i don't know how can I change the channel tranmission wifi
> in diferents APs... I read in another post, that this option is only

What is the "channel transmission wifi" ?

Mathieu

[Lluís Salvador Falumí]

Sorry Matieu,

I undestand channel transmision wifi how the diferences frequency that works APs. The channel is the frecuency: Channel 1, 2, 3,... to 11. Do you know, if I work with two AP, what is the error bytes (if the two APs are transmision another station) between them?

Thanks

2009/3/31 Mathieu Lacage <mathieu...@sophia.inria.fr>

[Mathieu Lacage]

On Tue, 2009-03-31 at 15:48 +0200, Lluís Salvador Falumí wrote:
> Sorry Matieu,
>
> I undestand channel transmision wifi how the diferences frequency that
> works APs. The channel is the frecuency: Channel 1, 2, 3,... to 11. Do
> you know, if I work with two AP, what is the error bytes (if the two
> APs are transmision another station) between them?

There is no support to modelize cross-channel interference. Nicola Baldo
said he was working on something to do this.

Mathieu

[Lluís Salvador Falumí]

OK.

From your paper [1] we had understood that inter-cell interference was
implemented but now I realize from eq. 3 that "Ni(k, t) represents the
interference noise, that is, the sum of the energy of all the other
signals received on the SAME CHANNEL".

It means that, at least, co-channel interference is working in ns-3(?).
However, the idea of including the effects of interference from other
channels would follow the same steps, but this time adding attenuation to
the interfering packets, depending on the channel distance between the
receiver and the interferer. These filtering for 802.11b and 11a/g was
computed in [2].
If we want to add this effect, which source files should we look first?

[1]: Mathieu Lacage and Thomas R. Henderson. Yet another network
simulator. In WNS2'06: Proceeding from the 2006 workshop on ns-2: the IP
network simulator, 2006.

[2]: E. Garcia-Villegas, E. López, R. Vidal, and J. Paradells. Effect of
adjacent-channel interference in IEEE 802.11 WLANs. In 2nd Int. Conference
on Cognitive Radio Oriented Wireless Networks and Communications,
CrownCom'07, pages 118-125, August 2007.

Thanks

2009/4/1 Mathieu Lacage <mathieu...@sophia.inria.fr>

[Nicola Baldo]

Hi,

> From your paper [1] we had understood that inter-cell interference was
> implemented but now I realize from eq. 3 that "Ni(k, t) represents the
> interference noise, that is, the sum of the energy of all the other
> signals received on the SAME CHANNEL".
>
> It means that, at least, co-channel interference is working in ns-3(?).
> However, the idea of including the effects of interference from other
> channels would follow the same steps, but this time adding attenuation to
> the interfering packets, depending on the channel distance between the
> receiver and the interferer. These filtering for 802.11b and 11a/g was
> computed in [2].
> If we want to add this effect, which source files should we look first?

Well, definitely you should have a look at the base classes in wifi-
channel, wifi-phy, and their implementation in yans-wifi-
{phy,channel}.
The point is that the current wifi phy does not have any notion of the
frequency it is tuned to, so in order to implement adjacent channel
interference this notion needs to be added somewhere.

On ns-dev we discussed about some possible ways to implement this
feature, see these threads:

http://www.nabble.com/spectrum-modeling-and-wifi-device-tt22183189.html#a22183189
http://www.nabble.com/Python-Bindings-for-optional-ns-3-features-tt22623327.html#a22629833

(you will have to dig a little bit into those threads, as we discuss
many other things which are not necessarily releated...)

I am currently working on the Spectrum modeling framework I mentioned
in the first thread, but I still have to integrate it with
ns3::WifiPhy

Regards,

Nicola