How to know to which enb the ltedevice is connected?

497 views
Skip to first unread message

Filipe Ferreira

unread,
May 27, 2014, 8:37:29 AM5/27/14
to ns-3-...@googlegroups.com
Hi, I'm working with cell zooming techniques for lte-advance and I'm using ns3 to verify the energy efficiency that I can get with them. I need to know the number of UEs that are connected to the differents enbs and the Id of which UE that is connected to every enb. I'm using the automatic Attach and I was trying to know the enb that the UEs are connected by getting the GetTargetEnb(), but when I verified the code of the function Attach, the SetTargetEnb() is never used. There is another way to get this information?

Regards,
Filipe Ferreira 

Anthony Faustine

unread,
May 27, 2014, 10:12:46 AM5/27/14
to ns-3-...@googlegroups.com
Help how do you measure energy on LTE module?
Can you share 



Anthony Faustine
Assistant Lecturer University of Dodoma..
Contact: +255752995506, (skype) sambaiga1


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

Filipe Ferreira

unread,
May 27, 2014, 7:11:36 PM5/27/14
to ns-3-...@googlegroups.com
Well, I will do it manually, because the module isn't prepared to that. I will have to write the times that I will put the enb in sleep.

Tiago Cerqueira

unread,
May 27, 2014, 8:05:33 PM5/27/14
to ns-3-...@googlegroups.com
I believe you can get that information by using the ASCII traces and checking the generated files

Filipe Ferreira

unread,
Jun 2, 2014, 11:14:56 AM6/2/14
to ns-3-...@googlegroups.com
Can you give me any help with that? I tried to "enable" the AsciiTraceHelperForDevice, but I'm not being successfull.

preeti.do...@gmail.com

unread,
Jun 3, 2014, 1:57:23 AM6/3/14
to ns-3-...@googlegroups.com
Hello,

:) I see someone finally working on Sleep Modes.. I am trying to achieve it on LTE.
I am using dual-strip-lena example. Added the codes for callbacks and getting the RSRP values for the surrounding cells..

Am stuck at the same position. Finding the number of UEs attached to every eNB. I need to handover these to the surrounding cells based on second-best-RSRP values..

Regards,
Preeti

Filipe Ferreira

unread,
Jun 3, 2014, 3:20:32 PM6/3/14
to ns-3-...@googlegroups.com
I'm just in the beginning, for what I was reading, the ns3 doesn't support a sleep mode switch, the answer I found (in another thread) was to lower the txpower of the cell (eg. -60). I tested it and checked the REM file with the cell area coverage and it seems to work. I'm using the x2-handover and as soon I decreased the txpower, 0.2 secs later the handover of the UEs was made automatically. 

Now I'm stuck like you, because I didn't find a way to limit the number of UEs each eNB can handle and I'm thinking to do it with auxiliar classes.

Filipe Ferreira

unread,
Jun 3, 2014, 3:27:15 PM6/3/14
to ns-3-...@googlegroups.com
Btw, I'm having some trouble to find the answer to the following questions:

- For example, if the eNB is working to a coverage area of 200 meters, how many users can it handle? If I increase the coverage area, what is the impact on the number of users?

preeti.do...@gmail.com

unread,
Jun 4, 2014, 7:43:17 AM6/4/14
to ns-3-...@googlegroups.com
yes. I also need to use auxilary classes. 

If you are using automatic attach to eNB function, I think you have very little control over the max threshold for the number of UEs that can connect to a cell.

Reg the sleep modes, I am only setting the Tx to a lower power value as the post says. This should do.

Filipe Ferreira

unread,
Jun 4, 2014, 9:57:18 AM6/4/14
to ns-3-...@googlegroups.com
I checked some of the examples I found like lena-dual-stripe and I think we can get the information about the UE->eNB. I'm "overwritting" NotifyConnectionEstablishedUe (to know the first attach) and NotifyHandoverEndOkUe (to know the new eNB). Both methods have the imsi of the UE and the CellId, both are independent to every UE and eNB, we just have to add some code lines to save the information on auxiliary classes.

Tiago Cerqueira

unread,
Jun 4, 2014, 10:34:51 AM6/4/14
to ns-3-...@googlegroups.com
I don't know if you've found out about this already, but you need to enable PhyTraces. You can do that by calling the EnablePhyTraces() function from the LteHelper object.
I'm doing this at the end of the simulation (before calling Simulator::Stop()), but I don't know if it will still work if you enable it at any point of the sim...

preeti.do...@gmail.com

unread,
Jun 4, 2014, 10:48:14 AM6/4/14
to ns-3-...@googlegroups.com
Yes.. Am using a Struct and a Map inexed by IMSI..Something like this am doing..


    struct UeMeas
  {
    uint64_t imsi;
    uint16_t cellId;
    uint16_t rnti;
    double rsrp;
    double rsrq;
  };
 
    std::map <uint64_t, UeMeas> UeMeasMap;

void
NotifyConnectionEstablishedEnb (std::string context,
                                uint64_t imsi,
                                uint16_t cellid,
                                uint16_t rnti)
{
   
    //newEl.servingCell = false;
  std::cout << Simulator::Now ().GetSeconds () << " " << context
            << " eNB CellId " << cellid
            << ": successful connection of UE with IMSI " << imsi
            << " RNTI " << rnti
            << std::endl;
            //if(cellid>=9 && cellid<=11){ // For the desired cell or capture all
                if(cellid==2){
            UeMeas newEl;
        newEl.rnti = rnti;
        newEl.imsi = imsi;
        newEl.cellId = cellid;
        newEl.rsrp = 0; // Initial Values. Will Update them when I call ReportUEMeasurements
        newEl.rsrq = 0; // Initial Values
            UeMeasMap.insert (std::pair <uint16_t, UeMeas> (imsi, newEl));

preeti.do...@gmail.com

unread,
Jun 4, 2014, 10:51:23 AM6/4/14
to ns-3-...@googlegroups.com
Hey Tiago,

We actually need to store these values in some structure and use them later in the simulation for processing.

Phy traces print the dump onto a text file. But using a callback to a trace, we have more power in terms handling and storing the
statistics.. So I guess, We are both using the auxillary data structures


Regards,
Preeti..

Filipe Ferreira

unread,
Jun 4, 2014, 2:59:58 PM6/4/14
to ns-3-...@googlegroups.com
In my case, I just really want to know the cellid of the enb, so, for now, I'm using a simple data structure. I will let the automatic handover choose the new best cell for the ue.

preeti.do...@gmail.com

unread,
Jun 17, 2014, 2:19:24 AM6/17/14
to ns-3-...@googlegroups.com
Hello,

In your script, do u have a requirement that requires to add UEs during run time.
I mean, in dual lena example, depending on the density of the UEs, the number if UEs are precalculated,

In my simulation, I need to introduce some more UEs, after simulation has started. Could u pls comment on this.

BR,
Preeti

Filipe Ferreira

unread,
Jun 17, 2014, 3:57:14 PM6/17/14
to ns-3-...@googlegroups.com
I worked on it today. I didn't research too much about this, but a simple way to do it is to do it manually using handlers. You have to create all the nodes, install the ip stack and install ueltedevice. With the handler, just make the attachment with the eNB. If you wanna do this in a random way, use random functions for the time of the handler and the number of ues.

Regards,
Filipe Ferreira.

preeti.do...@gmail.com

unread,
Jun 18, 2014, 9:28:09 AM6/18/14
to ns-3-...@googlegroups.com
Do you You mean to say that I simply schedule a function at a random time. That function should contain the attach statement.
Let me try it.  Thanks you.

T and BR,
Preeti

Filipe Ferreira

unread,
Jun 18, 2014, 3:55:25 PM6/18/14
to ns-3-...@googlegroups.com
Yes, but you need to create all the nodes, ip stack, etc.. in the beginning. If you don't do that it will got an error: "Cannot create UEs in real time" or something like that.

preeti.do...@gmail.com

unread,
Jun 20, 2014, 6:35:10 AM6/20/14
to ns-3-...@googlegroups.com
Did you manage to find the amount of energy saved during the cell-off times? 

Preeti.

Filipe Ferreira

unread,
Jun 20, 2014, 9:15:12 AM6/20/14
to ns-3-...@googlegroups.com
I'm still in tests, so I don't have any results. With the bibliography I red, researches say that it's difficult to give a exact value, but they usually use 1600W /hour (if memory serves me right). I think I will make the amount energy saved calculations by "hand" (outside ns3). 

preeti.do...@gmail.com

unread,
Jun 20, 2014, 9:20:57 AM6/20/14
to ns-3-...@googlegroups.com
If u give me ur mail Id, I will mail u a project on ns3 that has implemented energy module for LTE.
It may help u.

And btw, which propagation loss model are you using. 

Anthony Faustine

unread,
Jun 20, 2014, 9:39:38 AM6/20/14
to ns-3-...@googlegroups.com

samb...@gmail.com

Send me that lte energy model

preeti.do...@gmail.com

unread,
Jun 20, 2014, 10:04:37 AM6/20/14
to ns-3-...@googlegroups.com

I found it when I was scouting for pointers for my project. I still have to read through it to understand its contents. 
Pls have a look at it too. 
To unsubscribe from this group and stop receiving emails from it, send an email to ns-3-users+unsubscribe@googlegroups.com.

Anthony Faustine

unread,
Jun 20, 2014, 10:21:32 AM6/20/14
to ns-3-...@googlegroups.com

Thank you i will check on it

Anthony Faustine

unread,
Jun 21, 2014, 2:38:02 PM6/21/14
to ns-3-...@googlegroups.com
I had passed through the doc

all the required helper and model .cc codes are included but it lacks .h codes...
Any one with this code can share..



Anthony Faustine
Assistant Lecturer University of Dodoma..
Contact: +255752995506, (skype) sambaiga1


Reply all
Reply to author
Forward
0 new messages