Fading module functioning

[Crisan Sergiu]

Hi everybody,

I want to run a simulation using the LENA module for LTE with multiples number of users (both stationary and moving) and eNBs, to test the performance of different frequency reuse algorithms implemented in ns-3. In this simulation i want to add besides the propagation loss a fading component. I read the information available in the model library about fading and i did not find a lot of information  about the fading module, only general purpose information. My questions about the fading module are the following: 

1)Is it possible to add simultaneously two different fading files in the simulation, one for the downlink channel and one for the uplink channel ?

2)Is it possible to configure somehow the scenario so that only certain UEs are affected by fading ?

3) Does a fading trace file created for a speed of 45 kmph affect stationary UEs in the same simulation, or just the UEs which are moving with that speed ?

4)If I add in the simulation simultaneously multiple different fading trace file for different speeds, does the fading affect only the users which are moving with that speed ?

Thanks in advance

P.S : I am using ns-3 version 3.29

[Biljana Bojović]

Hi Crisan, 

On Wednesday, May 13, 2020 at 4:30:50 PM UTC+2, Crisan Sergiu wrote:

Hi everybody,

I want to run a simulation using the LENA module for LTE with multiples number of users (both stationary and moving) and eNBs, to test the performance of different frequency reuse algorithms implemented in ns-3. In this simulation i want to add besides the propagation loss a fading component. I read the information available in the model library about fading and i did not find a lot of information  about the fading module, only general purpose information. My questions about the fading module are the following: 

1)Is it possible to add simultaneously two different fading files in the simulation, one for the downlink channel and one for the uplink channel ?

Yes, there could be some improvements. Related to your question if you go into lte-helper.cc you will find the following code:

  if (!m_fadingModelType.empty ())

    {

      m_fadingModel = m_fadingModelFactory.Create<SpectrumPropagationLossModel> ();

      m_fadingModel->Initialize ();

      m_downlinkChannel->AddSpectrumPropagationLossModel (m_fadingModel);

      m_uplinkChannel->AddSpectrumPropagationLossModel (m_fadingModel);

    }

So, this says that the same fading model is set for downlink and uplink channel. I think that it would not be too difficult to split this into two parts, so all functions that were configuring a unique fading model to be converted into two functions: one for DL and another for UL fading model. 

This should be very quick and easy to do. And that kind of enhancement could be a candidate to be included in ns-3-dev.

 

2)Is it possible to configure somehow the scenario so that only certain UEs are affected by fading ?

No. If you would like to see the class go to spectrum/trace-fading-loss-model.h/cc. Maybe you could add there some new logic, but this does not look so straightforward.

 

3) Does a fading trace file created for a speed of 45 kmph affect stationary UEs in the same simulation, or just the UEs which are moving with that speed ?

Very good question. Hm.. I think that now I understand better your question 2. So, again, the code is in trace-fading-loss-model.h/cc and I dont see there any distinction between stationary/mobile UEs. Also, as we see in answer for question 1. in LteHelper the fadingModel is assigned for the channel, so this function DoCalcRxPowerSpectralDensity will be called for all UEs. 

If the distinction you want to make is based on speed, I think that yes, you could add quite simple tweak in trace-fading-loss-model.cc, which will check the velocity of nodes and if they are stationary it will just not apply the fading, it would just return from the function txPsd.

4)If I add in the simulation simultaneously multiple different fading trace file for different speeds, does the fading affect only the users which are moving with that speed ?

So, again, the file that contains almost all your answers is trace-fading-loss-model.h/cc, yes, I think that you can extend this implementation to apply fading for UEs that have speed +- delta of the speed for which is generated the fading model (it will depend on the fading file). Yes, theoretically, you can chain various pathloss/spectrum models, and for each UE would be applied only the corresponding one based on speed. 

This could be also valuable contribution to ns-3 module.

  

Thanks in advance

P.S : I am using ns-3 version 3.29

 

Best regards, 

Biljana

[Crisan Sergiu]

Hi Biljana Bojović,

Thanks for the detailed answers to my questions. I will try to add the aforementioned changes to the fading module and if successful I will send the code back to the ns-3 developers to be checked and possibly added in future versions of the simulator.

--
Posting to this group should follow these guidelines https://www.nsnam.org/wiki/Ns-3-users-guidelines-for-posting
---
You received this message because you are subscribed to the Google Groups "ns-3-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ns-3-users+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/ns-3-users/ea50e6b3-c1f7-4a4d-9a03-57635bec93ed%40googlegroups.com.