[Ns-developers] NS-3 Wifi contributions

[Federico Maguolo]

Hi Mathieu
sorry for the delay, but I found a little bug on the WifiModeFactory that 
takes to me a few to be discovered.
At the usual link (http://www.dei.unipd.it/wdyn/index.php?IDsezione=5517) you 
can find all my patches. The three patches that I think you are interested 
are:
-Bug Fix for WifiMode: this is the fix for the problem that I mentioned above
-Propagation Loss Model Interface Change: this is the change to the interface 
of the PropagationLossModel as we discuss in the previous mail
-Jakes model for Wifi: the jakes and composite propagation loss models with 
the suggestions that you write to me in the previous mail and according to 
the new PropagationLossModel interface.

It will be a pleasure for me to drink a beer with you in Italy :-)

Federico

On Wednesday 19 March 2008 03:54, Mathieu Lacage wrote:
> On Mon, 2008-03-17 at 12:15 +0100, Federico Maguolo wrote:
> > > Now that ns-3-param has been merged into ns-3-dev, I have looked again
> > > at your Jakes and Composite propagation loss models. The patches look
> > > nice but before being integrated, they would need to be updated to the
> > > new API in ns-3-dev: do you think you could produce a new patch ?
> >
> > An updated patch for jakes and composite propagation loss models is ready
> > at
> >
> > http://www.dei.unipd.it/wdyn/index.php?IDsezione=5517
>
> The patch looks good, modulo a few comments:
>   - can you remove the commented-out includes at the top of the .cc
> file ?
>   - can you split the file in two files :
> composite-propagation-loss-model and jakes-propagation-loss-model ?
>   - can you try to add some doxygen doc to your new classes ? i.e., you
> need to add /** */ blocks at the top of each class together with a
> \brief tag. There are some examples of this in src/core/object.h.
>
> > > Another thing I was wondering about is whether we could rename
> > > PropagationLossModel::GetRxPower (double,Ptr<MobilityModel> a,
> > > Ptr<MobilityModel> b) to PropagationLossModel::GetLoss (
> > > Ptr<MobilityModel> a, Ptr<MobilityModel> b) ? The latter would more
> > > closely match the behavior of PropagationDelayModel which returns the
> > > 'delay' rather than the time of expiration but I wonder if it would be
> > > enough to express the semantics of a sufficient number of propagation
> > > loss models. What do you think ?
> >
> > Yes, the propagation loss is a coefficient which is independent from the
> > transmit power, thus I think you can change the interface as you suggest.
>
> If you could also send another patch to do that, you would become my
> hero of the day/week (month ?) and I would owe you yet another bear
> [and/or beverage of your choice] the next time we meet in italy :)
>
> [I hope it does not show too much that I am trying to get you to do my
> job :)]
>
> thanks again for your efforts,
> Mathieu