The simulation of satellite

[lyf....@gmail.com]

I am a new scholar of NS3. There is not a satellite module related to NS3. I am doing a simulation in heterogeneous network, which contains the satellite network. But I do not know how to carry out. Hope you can give me some advice on how to achieve the satellite network simulation. I consider to realize the heterogeneous communication in the range of one satellite, which which contains LTE and WiFi. 

Is there anyone has used NS3 to do a simulation of satellite? Is there any public satellite module? Hope you can help me. Thank you.

[Tommaso Pecorella]

There is no public satellite module yet.

There is one, but it's not public. As far as I know it will be released when the IPR issues will be fixed. Due to some delays and bureaucratic craziness, it's not yet released.

Cheers,

T.

[lyf....@gmail.com]

Thank you. I have a few questions.

I want to complete a simple satellite module which can achieve the communication between a satellite and a base station of LTE. I don't know whether the simulation is complicated.

If I set the configuration of a satellite like the base station of LTE, in which just setting the parameters of the signal gain and data rate  and so on. Is it feasible?

在 2016年9月2日星期五 UTC+8上午6:57:43，Tommaso Pecorella写道：

[Tommaso Pecorella]

It is feasible. Meaning, you can approximate a bent-pipe satellite link with a low rate, high delay point to point channel.

However, the model is extremely rough. More or less like simulating a Tesla car with a soapbox car: they both have 4 wheels and neither has a combustion engine. The similarity stops here.

Cheers,

T.

[lyf....@gmail.com]

Thank you very much. 

Because I just started to learn NS3, I am not very familiar with the simulation. If I hope this simulation to be more credible, what factors need to be considered? Or do I need to simulate  some other modules? 

在 2016年9月3日星期六 UTC+8上午12:22:10，Tommaso Pecorella写道：

[Tommaso Pecorella]

I'd say (in order of priority):

- Delay

- Bandwidth assignment / MAC

- Frame structure (i.e., payload length)

- Channel model

Note that all the above could be fixed OR time-varying, depending on the satellite.

T.

[Natale Patriciello]

On 02/09/16 at 09:22am, Tommaso Pecorella wrote:
> It is feasible. Meaning, you can approximate a bent-pipe satellite link
> with a low rate, high delay point to point channel.
> However, the model is extremely rough. More or less like simulating a Tesla
> car with a soapbox car: they both have 4 wheels and neither has a
> combustion engine. The similarity stops here.

Hi Tommaso,

just a note: after many years of SOCIS we still don't have
a good model for a satellite link.

What I miss (and seems that the author misses it too) is a model that
does not consider only the link (losses, quality, whatever) but the
problem of accessing that channel: I mean, the contention phase.

I consider it very important because, while the channel can be easily
approximated (as you said, with a p2p) providing also close results to
the reality, with the use of a p2p channel you deliberately ignore the
contention/sharing part, which is (what an unlucky situation) the part
that differs more from the reality, diverging from it.

There's still plenty of publications that use such approximation (and
mines are one of them, unfortunately), but I consider it a flawed model.

Nat

[lyf....@gmail.com]

Thank you very much. 

I found the way to set the time delay and rate error model  in PointToPointHelper. I set these parameters to a fixed value. There are still some problems.

lteHelper->SetSchedulerType ("ns3::PfFfMacScheduler");

LTE sets scheduling method through this statement. Does PointToPointHelper have a similar setting statement? Could you give me some advice on how to set up the frame structure and the channel model?

dlClient.SetAttribute ("Interval", TimeValue (MilliSeconds (interPacketInterval)));

lteHelper->SetAttribute ("FadingModel", StringValue ("ns3::TraceFadingLossModel"));

I found these setting statements that sets the packets interval for the application and the LTE fading channel model. Can these statements be used in the PointToPointHelper?

在 2016年9月12日星期一 UTC+8下午3:56:36，Tommaso Pecorella写道：

[Tommaso Pecorella]

Hi,

no, these are LTE-specific.

P2P can use the new TrafficControl to prioritize packets, but (as Nat said) there's no dynamic bandwidth allocation mechanism in P2P.

About the channel model, it's to be defined anew, the terrestrial ones are (usually) not adequate for satellite links. Moreover, it really depends on the kind of terminal you need to simulate. A fixed terminal is pretty much always in good link state, while a mobile isn't.

Cheers,

T.

[Leonardo Castillo]

Hello

Is there anyone has used NS3 to do a simulation of satellite? Is there any public satellite module?

thanks