[Ns-developers] access point selection based on the received signal strength

ichrak amdouni ichrak.amdouni at gmail.com
Wed Jan 21 07:37:13 PST 2009


2009/1/21 Mathieu Lacage <mathieu.lacage at sophia.inria.fr>

> On Tue, 2009-01-20 at 14:14 +0100, ichrak amdouni wrote:
>
> >  Concerning the detection of disassociation, it is controlled by the
> > NqstaWifiMac::RestartBeaconWatchdog and NqstaWifiMac::MissedBeacons
> > method.
> > Here, nothing says that it is the access point to which the station is
> > associated is not sending beacons anymore, I expected to find
> > something that
> > keep track of that particular access point. So how can we set the
> > station
>


>
> When the station receives a beacon from the right access point (the one
> whose address matches the station's BSSID), it updates the beacon
> watchdog so, when the beacon watchdog expires, we know for sure that we
> have missed n consecutive beacons from the access point we were
> associated with.


Hi all,

Thanks to Mathieu.

I am working with ns3.2 and here the station restarts the beacon watchdog
whenever it receives a beacon even if it is not from the AP it is associated
to. Problem fixed, I add the relevant test on the sender address in my
version (as added in ns3.3).

After a first research and comparison of the two versions, I think nothing
related to the behavior of the Nqsta, Nqap, WifiPhy and WifiMac could lead
me to strange results as far as an access point selection application is
concerned (association, rate control...).

 Any suggestions will be very welcome.


>
>
> > state to beacon_missed and trigger probe requests' sending?
> >
>
> I don't really understand this question, sorry.


I mean that this station behavior ( to set state to beacon_missed and
trigger probe requests' sending ) is strange since it does not concern the
AP the station is associated to. Now the problem is fixed with the added
test.


As I said, I want to simulate the wifi association based on the received
signal strength.

       1. In the presence of two access points, a station receives beacons
having the same power even it is moving at high speed.

   1.

   It seems that all nodes are having the same WifiPhy (the same TX *power*,
   same CS
   threshold ...) . If it is the case, is this fact explaining 1) or it is
   perhaps related to the mobility model of the station?
   2.

   One way to make AP send with different txpower is to modify the txgain of
   the underlying physical layer. Right? Is there any other possibility?



Thank you in advance.
Ichrak

>
>
>
>


More information about the Ns-developers mailing list