Friis Propagation Loss model

[patriciaruiz]

Hi all,

I am working with the propagation loss models, testing most of them
and I have problems with Friis Propagation loss model. It is dropping
all of my packets even when I have only 3 devices and they are
separated only 10m each (and for tests devices don't move). With all
the other models I have no problems. Is there any error or any reason
why it works like that?

Thanks!

[patriciaruiz]

On 29 ene, 10:39, patriciaruiz <patriciaruizvillalo...@hotmail.com>
wrote:

In fact there is something I don't really understand. I have only one
propagation loss model but when calling double
PropagationLossModel::CalcRxPower (double txPowerDbm,
Ptr<MobilityModel> a,
Ptr<MobilityModel> b) const
{
double self = DoCalcRxPower (txPowerDbm, a, b);

if (m_next != 0)
{
self = m_next->CalcRxPower (self, a, b);
}
return self;
}

from the yans-wifi-channel , but I don't understand why it is entering
in the if and calling againg CalcRxPower with the power obtained the
first time.....

[patriciaruiz]

On 29 ene, 11:42, patriciaruiz <patriciaruizvillalo...@hotmail.com>
wrote:

I think is just because by default is set the logPropagationLossModel
and when I add the friisPropagationLossModel I have then two
propagation model instead of one.
Is that the possible error??

[R G]

Hi patriciaruiz!

I'm new with ns-3 but in description here: http://www.nsnam.org/docs/release/3.17/doxygen/classns3_1_1_friis_propagation_loss_model.html#details
written that this model is only for long distance:
 "This model is invalid for small distance values. The current implementation returns the txpower as the rxpower for any distance smaller than MinDistance."
If I understood smth wrong, sorry for useless advice.

[Konstantinos]

You can change this value if needed. Currently this MinDiastnace is 0.5meters.

Since distance is in the denominator, it can not be 0. If your inter-node distance is less than 0.5 then you might consider another model (e.g. LogDistance) which does not have any limitation.