Question regarding Dedicated Eps Bearer and handover

[Sahawat Namwanta]

Hello,

I have a question as follows

1. I'm now trying to have a 3 bearers among UEs. Is the following code the correct way to do so? (don't mind "%3", since i use it to divide the UEs into 3 group)

for (uint16_t i = 0; i < numberOfUes; i++)
        {
  if(i%3==0)
  {
            lteHelper->ActivateDedicatedEpsBearer(ueLteDevs.Get(i), VIDbearer, tftVIDEO);        
  }
  if(i%3==1)
  {  
            lteHelper->ActivateDedicatedEpsBearer(ueLteDevs.Get(i), VoIPbearer, tftVoIP);          
  }
  if(i%3==2)
  {  
            lteHelper->ActivateDedicatedEpsBearer(ueLteDevs.Get(i), Databearer, tftTCP);          
  }
 }

2. I'm noticing that, after handover occurs, the LteHelper seems to revert the dedicated bearer back to the default bearer. So, I take a look in lte-helper.cc and tried to remove both line 751 and 781

line 751: m_epcHelper->ActivateEpsBearer (ueDevice, ueLteDevice->GetImsi (),
                                  EpcTft::Default (),
                                  EpsBearer (EpsBearer::NGBR_VIDEO_TCP_DEFAULT));

line 781: m_epcHelper->ActivateEpsBearer (ueDevice, ueLteDevice->GetImsi (), EpcTft::Default (), EpsBearer (EpsBearer::NGBR_VIDEO_TCP_DEFAULT));

However, this does not seems to work. So, my question is if there are anyways to keep the dedicated bearer working even after the HO procedure? It will be very useful for some scheduler like CQA-PF (or any with QoS related).

thank you in advance everyone  :)

[Sahawat Namwanta]

Here is my code. I use CQA scheduler on my scenario involving handovers.

If you try logging the scheduler file, you'll see that when the UEs first attach to the first eNB, the scheduler is able to differentiate between bearers just fine. 

However, after the first handover, the scheduler were not able to differentiate between bearers and only see every bearers as the default bearers (EpsBearer::NGBR_VIDEO_TCP_DEFAULT).

Any help would be appreciated.  :)

Thank you so much in advance.

[Sahawat Namwanta]

Hello again !!

I tried changing the handover algorithm to 

  lteHelper->SetHandoverAlgorithmType ("ns3::NoOpHandoverAlgorithm");

and make it handover at t = 4s

 lteHelper->HandoverRequest (Seconds (4), ueLteDevs.Get (u), enbLteDevs.Get (0), enbLteDevs.Get (1));

then i added the debug logging in CqaFfMacScheduler

as you can see, after i ran the code with 1 UE

export 'NS_LOG=CqaFfMacScheduler=level_debug|prefix_time'

./waf --run "scratch/train_cqa --numberOfUes=1"

the Video traffic starts around t = 2s (fig.1)

fig.1 The scheduler sees the bearer as Video

then  at t = 4s the handover occurs and the traffic revert itself back to the default bearer!! (fig.2)

fig.2 after the handover, the bearer seems to revert itself back to default !!

is this a known issue? I was stuck with this problem for a while now. Any answers would be super appreciate!!

thanks

[Biljana Bojović]

Hi, 

sorry for delay in reply.  Yes, I think that this is a bug, I will take a look asap. Tom opened a bug https://www.nsnam.org/bugzilla/show_bug.cgi?id=2777. Please follow it.

Cheers,

Biljana

[Sahawat Namwanta]

Thank you sooo much. I thought it was me all along.

Cheeeeers !!

[Biljana Bojovic]

Hi Sahawat,

the results that you provided in these figures, do you have results for more time ater 4.002 seconds? I would like to see around 4.050-4300 seconds. I am running simulation but it takes to long.

Cheers,

Biljana

On Fri, Aug 25, 2017 at 6:25 PM, Sahawat Namwanta <s.nam...@gmail.com> wrote:

Thank you sooo much. I thought it was me all along.

Cheeeeers !!

--
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+unsubscribe@googlegroups.com.
To post to this group, send email to ns-3-...@googlegroups.com.
Visit this group at https://groups.google.com/group/ns-3-users.
For more options, visit https://groups.google.com/d/optout.

[Sahawat Namwanta]

Hello Biljana,

Sorry for the late reply. 

here's what happened to the whole the simulation.

(Note that I change the HO time to be 10s 20s 30s)

for(uint32_t u = 0; u < numberOfUes; ++u)
{
  lteHelper->HandoverRequest (Seconds (10), ueLteDevs.Get (u), enbLteDevs.Get (0), enbLteDevs.Get (1));
  lteHelper->HandoverRequest (Seconds (20), ueLteDevs.Get (u), enbLteDevs.Get (1), enbLteDevs.Get (2));
  lteHelper->HandoverRequest (Seconds (30), ueLteDevs.Get (u), enbLteDevs.Get (2), enbLteDevs.Get (3));
}

(Figure 1. simulate for 1 UE)

(Figure 2 The Video traffic starts)

(Figure 3 The first handover occurred)

(Figure 4 The first handover)

(Figure 5 The second handover)

(Figure 6 The third handover)

(Figure 7 The simulation finishes)

And the attach file is the whole simulation!

(Note that the timing of the traffic sequence is a bit different in the attach file, since I ran this simulation after the ones showing in the above figures and the seed of the random numbers are different) 

Please let me know if you need anything more and thank you so much for your kind attention

Cheers !, Sahawat.

On Saturday, 26 August 2017 00:06:11 UTC+7, Biljana Bojović wrote:

Hi Sahawat,

the results that you provided in these figures, do you have results for more time ater 4.002 seconds? I would like to see around 4.050-4300 seconds. I am running simulation but it takes to long.

Cheers,

Biljana

On Fri, Aug 25, 2017 at 6:25 PM, Sahawat Namwanta <s.nam...@gmail.com> wrote:

Thank you sooo much. I thought it was me all along.

Cheeeeers !!

--
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.

[Sahawat Namwanta]

Here's another simulation with 3 UEs at the beginning (Video VoIP and Data transmission)

Here --> bearer-handover-3.txt

(I have to change the logging in CqaFfMacScheduler in the attach file for a little to be able to display the VoIP traffic properly!)

(Figure 1  The first handover)

Best regards,

Sahawat.

[Sahawat Namwanta]

Also, I'm not sure that you really mean 4300 seconds or 4.300 seconds, but I conducted the 4300 seconds of simulation anyway. (But if you want to observe on the time when the handover is completed please have a look at the above post of the past 10 secs)

So here's the result of the 4300 secs of simulation -- link >> 1 UE 4300 secs (The file is quite big)

(Again, the the handover is started at  t = 10 s , 20 s, and 30 s)


Thanks again

- Sahawat

On Saturday, 26 August 2017 00:06:11 UTC+7, Biljana Bojović wrote:

Hi Sahawat,

the results that you provided in these figures, do you have results for more time ater 4.002 seconds? I would like to see around 4.050-4300 seconds. I am running simulation but it takes to long.

Cheers,

Biljana

On Fri, Aug 25, 2017 at 6:25 PM, Sahawat Namwanta <s.nam...@gmail.com> wrote:

Thank you sooo much. I thought it was me all along.

Cheeeeers !!

--
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.

[Biljana Bojovic]

Sorry, it was my typo. I wanted to see what happens until 4.300 seconds. Anyway, thanks for the traces. I will take a look.

BR,

Biljana

To unsubscribe from this group and stop receiving emails from it, send an email to ns-3-users+unsubscribe@googlegroups.com.

[Zoraze Ali]

Hi Sahawat,

I think I figured it out where is the problem. Could you please apply the attached patch and test.

Kind regards,
Zoraze
 

[Biljana Bojovic]

Zoraze thanks a lot for the proposed patch!

Sahawat can you please provide some feedback if this patch works for you.

Thanks!

Biljana

--

[Sahawat Namwanta]

@Zoraze, @Biljana

The patch fixes the problem !!

The bearer ID seems to stay the way it should.

Thank you so much Zoraze and Biljana.

sincerely

- Sahawat