[Ns-developers] Channel scanning in 802.11

[Giuseppe Barba]

Hi all,
in order to correctly simulate scenarios with multi channel communication,
you must model the adjacent channel interference (ACI) and the adjacent
channel rejection (ACR). For this reason is  necessary to preventively
engineering the Phy model.
For this scope we can just consider the ideal TX power masks of various
classes of TX filter (class A, B, C e D) to model the overlapping channel
(ACI), and the RX filter mask to consider the rejection to adjacent
interference with various classes of filter (e.g. class 2).
For example: if a node transmit over an adjacent channel, we can compute the
correspondent overlapped TX interfered power by attenuate the TX power by a
factor depending on Tx mask at frequency shift. Then we can compute the RX
interfering power through loss model adopted for in channel communication.

Regards,
Giuseppe

2008/10/10 Gustavo Carneiro <gjcarneiro at gmail.com>

> A while back I had written some wifi scanning code.  I have re-synced with
> ns-3-dev and uploaded the code here, in case it is useful to someone:
>
> http://code.nsnam.org/gjc/ns-3-wifi-scanning/
>
> Caveats:
>
>  1- The code is experimental;
>
>  2- The active scanning is pessimistic, i.e. scanning takes longer time
> than real world, because it assumes all channels are busy with traffic and
> always spends 50 ms in each channel (real world cards are smarter and spend
> just a few ms if they do not sense any traffic in that channel).  In any
> case, it works, in spite of being rudimentary.
>
>  3- The code has not been reviewed yet by Mathieu, the WiFi maintainer,
> although I hope he will consider it ;-)
>
>  4- I don't have a lot of time to maintain the code, so it would be
> wonderful if someone could pick it up.  Maybe I might develop something
> more
> on top of this, or maybe not, I am not sure yet...
>
> Further things to implement:
>
>  1- Switching channels should consume between 1 and 5 ms (hardware
> limitations) [1];
>
>  2- Probing a "free" channel should require a delay of only 1-2 ms (aka
> "minimum channel time"), and 10 to 27 ms (aka "maximum channel time") for
> busy channels [1];
>
>  3- Due to inter-channel interference, additional channels may erroneously
> appear busy and thus cause additional scanning delays (has to spend maximum
> channel time, instead of minimum, for additional channels), so without the
> inter-channel interference model mentioned by Mathieu scanning delays might
> be too optimistic;
>
>  4- The scanning machinery should somehow record the "signal quality" of
> the received ProbeResponses, and report it back to the client, so that the
> client can choose the best AP based on signal strength;
>
>  5- The final piece of the puzzle would be to implement a MobilityManager
> kind of object, attached to the node, which would select the best AP at all
> times.  Right now NS 3 has very rudimentary mobility management.  An
> experiment I have made: place two APs on the same channel, same SSID,
> spaced
> by 100 meters;  move a STA to the space in the middle of the two APs; the
> STA starts receiving beacons from both APs, and it keeps associating to one
> and other AP, constantly flip-flopping its association between the two APs.
>
> [1] Murray, D.<
> http://ieeexplore.ieee.org/search/searchresult.jsp?disp=cit&queryText=%28murray%20%20d.%3CIN%3Eau%29&valnm=Murray%2C+D.&reqloc%20=others&history=yes
> >
> Dixon, M.<
> http://ieeexplore.ieee.org/search/searchresult.jsp?disp=cit&queryText=%28%20dixon%20%20m.%3CIN%3Eau%29&valnm=+Dixon%2C+M.&reqloc%20=others&history=yes
> >
> Koziniec, T.<
> http://ieeexplore.ieee.org/search/searchresult.jsp?disp=cit&queryText=%28%20koziniec%20%20t.%3CIN%3Eau%29&valnm=+Koziniec%2C+T.&reqloc%20=others&history=yes
> >
> , "Scanning Delays in 802.11 Networks", NGMAST '07. (
> http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=4343430)
>
> 2008/10/7 Mathieu Lacage <mathieu.lacage at sophia.inria.fr>
>
> >
> > On Tue, 2008-10-07 at 04:23 -0700, Ramon Bauza wrote:
> > > Hi all,
> > >
> > > I have seen in the ns-3 manual, that although WiFi networks can be
> > > simulated in infraestructure mode, the current ns-3 version does not
> > > perfom channel scanning. My questions are then:
> > >
> > > - Can a NetDevice switch among different channels at execution time?
> >
> > You can change the frequency band used by a device but there is no
> > support in the current PHY model to modelize interference among signals
> > in overlapping frequency bands. A patch/model description to do so would
> > be welcome.
> >
> > > - Is the channel scanning going to be implemented?
> >
> > yes, if you send a patch.
> >
> > regards,
> > Mathieu
> >
> >
> > --~--~---------~--~----~------------~-------~--~----~
> > You received this message because you are subscribed to the Google Groups
> > "ns-3-users" group.
> > To post to this group, send email to ns-3-users at googlegroups.com
> > To unsubscribe from this group, send email to
> > ns-3-users+unsubscribe at googlegroups.com<ns-3-users%2Bunsubscribe at googlegroups.com>
> <ns-3-users%2Bunsubscribe at googlegroups.com<ns-3-users%252Bunsubscribe at googlegroups.com>
> >
> > For more options, visit this group at
> > http://groups.google.com/group/ns-3-users?hl=en
> > -~----------~----~----~----~------~----~------~--~---
> >
> >
>
>
> --
> Gustavo J. A. M. Carneiro
> INESC Porto, Telecommunications and Multimedia Unit
> "The universe is always one step beyond logic." -- Frank Herbert
>