Propagation loss, fading and shadowing

[Joao]

Dear all,

In the wifi-simple-adhoc-grid.cc example, the
FriisPropagationLossModel is added to wifiChannel. From my
understanding, this means we now have two propagation models: the
default LogDistancePropagationLossModel and the
FriisPropagationLossModel. Are both active simultaneously? What is the
use of this? Or is the FriisPropagationLossModel just added to
calculate the reference loss at the reference distance, as required by
the log-distance model?

Also, is it possible to use more elaborate fading models such as
Rayleigh fading and also shadowing with ns3?

Thanks,
João

[Tom Henderson]

On 01/03/2011 09:34 AM, Joao wrote:
> Dear all,
>
> In the wifi-simple-adhoc-grid.cc example, the
> FriisPropagationLossModel is added to wifiChannel. From my
> understanding, this means we now have two propagation models: the
> default LogDistancePropagationLossModel and the
> FriisPropagationLossModel. Are both active simultaneously? What is the
> use of this?

It is possible to chain propagation loss models together so that they
each act on the packet (such as a large scale plus a small scale fading
model).

The underlying method PropagationLossModel::SetNext() is documented as
follows:

* This method of chaining propagation loss models only works
commutatively
* if the propagation loss of all models in the chain are independent
* of transmit power.
*/
void SetNext (Ptr<PropagationLossModel> next);

Not all of the propagation loss models have this commutative property,
however.

Or is the FriisPropagationLossModel just added to
> calculate the reference loss at the reference distance, as required by
> the log-distance model?

In the example you cite (wifi-simple-adhoc-grid.cc), there is only the
Friis model.

The key here is how this helper object is created. Note that it is
created without any loss or propagation models:

YansWifiChannelHelper wifiChannel ;

If, however, it had been created as:
YansWifiChannelHelper wifiChannel = YansWifiChannelHelper::Default ();

then these would have been added by the Default() method:

helper.SetPropagationDelay ("ns3::ConstantSpeedPropagationDelayModel");
helper.AddPropagationLoss ("ns3::LogDistancePropagationLossModel");

and then Friis would have been subsequently added by the
wifi-simple-adhoc-grid program after LogDistance.

>
> Also, is it possible to use more elaborate fading models such as
> Rayleigh fading and also shadowing with ns3?
>

Yes, see the file propagation-loss-model.h for the available models
(plus there are a few additional ones like Jakes that are separately
declared in the src/common directory).

- Tom

[Alvin SW]

I could not find the Rayleigh fading and/or shadowing model in the
documentation.
There are only: Cost231PropagationLossModel,
JakesPropagationLossModel, RandomPropagationLossModel,
FriisPropagationLossModel, TwoRayGroundPropagationLossModel,
LogDistancePropagationLossModel, ThreeLogDistancePropagationLossModel,
NakagamiPropagationLossModel, FixedRssLossModel,
MatrixPropagationLossModel, and RangePropagationLossModel.

I think there is not enough documentation on those models for the
beginners. For example, the doxygen of Cost231PropagationLossModel
does not contain any description at all, what is it about, why is it
needed, references to underlying algorithms, how to use it, etc. Would
it be possible for each models writers to add a bit of information on
their models?

Regards,

Alvin

[Rafael Fernandes Lopes]

I don't know if this is applicable for the NS-3 implementation, but the Rayleigh fading channel is equal to the Nakagami fading channel with the parameter "m = 1". However, the NakagamiPropagationLossModel has different "m" values for different distances. A possible approach is define the attributes "m0", "m1" and "m2" equal to "0".

Any other ideas?

Regards,

--
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-...@googlegroups.com.
To unsubscribe from this group, send email to ns-3-users+...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/ns-3-users?hl=en.

--
Rafael Fernandes Lopes
Lattes:  http://lattes.cnpq.br/1972734433460838
____________________________________________________________
Universidade Federal de Campina Grande -- UFCG
Unidade Acadêmica de Engenharia Elétrica -- UAEE
Instituto de Estudos Avançados em Comunicações -- IECOM
http://www.iecom.org.br

Instituto Federal de Educação, Ciência e Tecnologia do Maranhão -- IFMA
http://www.ifma.edu.br
____________________________________________________________

[Joao]

Hello.

In the documentation for the Nakagami model, it says that if m=1 than
we have Rayleigh fading. Therefore, I suppose the m's should all be
set to 1 and not 0.

Regarding a shadowing model, there is something online but it does not
seem to have a parallel in terms of code.
http://www.nsnam.org/doxygen/classns3_1_1_shadowing_loss_model.html

Does someone know if there is shadowing model implemented or not?

Thank you,
Joao

On Jan 5, 3:11 pm, Rafael Fernandes Lopes <rafael.fernan...@gmail.com>
wrote:

> > ns-3-users+...@googlegroups.com<ns-3-users%2Bunsu...@googlegroups.com>
> > .

> > For more options, visit this group at
> >http://groups.google.com/group/ns-3-users?hl=en.
>
> --
> Rafael Fernandes Lopes
> Lattes:  http://lattes.cnpq.br/1972734433460838
> ____________________________________________________________
> Universidade Federal de Campina Grande -- UFCG
> Unidade Acadêmica de Engenharia Elétrica -- UAEE

> Instituto de Estudos Avançados em Comunicações -- IECOMhttp://www.iecom.org.br

[Rafael Fernandes Lopes]

Ops... Sorry, that was a typo mistake: as I said before (i.e., in the same paragraph), "m" must be set to 1! I'm sorry!!! :D

To unsubscribe from this group, send email to ns-3-users+...@googlegroups.com.

For more options, visit this group at http://groups.google.com/group/ns-3-users?hl=en.

[Joao]

Summing up:

- to include Rayleigh fading, it is sufficient to use the Nakagami
fading model with parameter m=1;
- Shadowing can be included by using the random propagation loss model
with, for example, a normal distribution with (mean,variance) as
parameters.

The values for mean and variance can be obtained in standard wireless
communication books.

João

On Jan 10, 7:11 pm, Rafael Fernandes Lopes

> > <ns-3-users%2Bunsu...@googlegroups.com<ns-3-users%252Buns...@googlegroups.com>

[Preethi Chandur]

My apologies, if my query seems redundant, but I just want some clarification regarding the files shadowing-loss-model.cc and shadowing-loss-model.h. As Joao asked previously, there seems to be documentation available for the same, but the files are not present in 3.14 nor in 3.15. I found some patches by Nicolo though and the report of a bug previously.
I am in need of the files since I am using log distance propagation loss model and I need to include the shadowing effects as a part of setting up 3GPP Case 1 scenario. It would be of great help if someone could let me know.

Thanks!!

> > <ns-3-users%2...@googlegroups.com<ns-3-users%252Bunsubscribe@googlegroups.com>

[Marco Miozzo]

Hi,

the BuildingsPropagationLossModel includes shadowing and propagation
loss models typical of 3GPP evaluation scenarios, too.
You can find more information in its documentation.

Best regards,
marco.

>> > > <ns-3-users%2...@googlegroups.com<ns-3-users%252Buns...@googlegroups.com>

> --
> You received this message because you are subscribed to the Google Groups
> "ns-3-users" group.

> To view this discussion on the web visit
> https://groups.google.com/d/msg/ns-3-users/-/X_jQ9soIEAoJ.

> To post to this group, send email to ns-3-...@googlegroups.com.
> To unsubscribe from this group, send email to

> ns-3-users+...@googlegroups.com.