LTE Carrier Aggregation

[Piotr]

Hi,

Where the source code of the GSOC2015LTECA is stored? I have downloaded the code from http://code.nsnam.org/daniloa/ns-3-dev-lte-ca and it turned out to be quite buggy, I'm afraid.

BR,

[Marco Miozzo]

Hi Piotr,

the code of CA is stored in the repository you indicated. Unfortunately, the project was very ambitious and the GSOC student has not been able to close the project with all the necessary  tests and therefore it has not been properly reviewed. We will be working on this during the next months and we hope we will be able to officially release the framework extended (with refined Carriers Manager) and validated (with examples and tests) at the end of 2016 approximatively.
In the meanwhile, if you are aware of any bug, it would be very helpful for us if you can report them, in order to take them into account.

Best,
marco.

--
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 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.

[Piotr Lechowicz]

Hi Marco and thanks for you reply!

In fact I will be glad if I could help with the development. Let me know if I could be of any assistance.

The things I noticed so far were:

1) lte-ue-net-device.c 

missing:

    .AddAttribute ("LteUeComponentCarrierManager",

                   "The CC Manager associated to this UeNetDevice",

                   PointerValue (),

                   MakePointerAccessor (&LteUeNetDevice::m_componentCarrierManager),

                   MakePointerChecker <LteUeComponentCarrierManager> ())

2) lte-helper.cc 

switched parameters:

    should be: 

DoCreateEnbComponentCarrierMap (dev->GetUlEarfcn (), dev->GetDlEarfcn (), dev->GetUlBandwidth (), dev->GetDlBandwidth ());

    and not: 

DoCreateEnbComponentCarrierMap (dev->GetUlEarfcn (), dev->GetDlEarfcn (), dev->GetDlBandwidth (), dev->GetUlBandwidth ());

In a couple of places the "for" loop is missing and instead a single assignment was used such as it = m_enbComponentCarrierMap.begin ();

3) lte-ue-rrc.cc and lte-ue-rrc.cc

stiff sizing to 1 in:

  m_cphySapUser.resize (1);

  m_cphySapProvider.resize (1);

  m_cmacSapUser.resize (1);

  m_ffrRrcSapUser.resize (1);

and some more, which I forgot already, I'm afraid.

Currently I'm trying to figure out why the 

LteUePhy::ReportRsReceivedPower (const SpectrumValue& power)

receives the empty power from time to time, but it still looks like a long way to go 🙁

BR,

Piotr

You received this message because you are subscribed to a topic in the Google Groups "ns-3-users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/ns-3-users/zv46lm6RSK0/unsubscribe.
To unsubscribe from this group and all its topics, send an email to ns-3-users+...@googlegroups.com.

[Szabolcs Nováczki]

Hi All,

I am also interested in the CA feature for LTE. To tell the truth I am quite new to ns-3, so i doubt I can make here useful contribution in the near future. Still i would like to get familiar with the implementation of CA. It would help me a lot, if i had an example script using the CA feature. Unfortunately, when cloning the repo, it turns out that there are no such scripts currently (however waf is looking for some, but the corresponding file is not available). Thus i asking for help here hoping you can provide an example script.

Reading through the discussion i cannot judge whether the CA implementation is in a shape where it is possible to run some simple simulations. If not then obviously the example that i am seeking wont be available. Still in this case, i would highly appreciate if you could point me in the right direction regarding how to start working with the CA feature.

BR,

Szabolcs

[Madan Pande]

Marco,
               There is one missing file "
I get the following error....I am compiling with waf after merging the lte module from lte-ca gsoc in to Lena
==============================================================
Waf: Entering directory `/home/ns3dev/repos/ns-3-allinone/lenaCA/build'
Waf: Leaving directory `/home/ns3dev/repos/ns-3-allinone/lenaCA/build'
source not found: 'lena-componentcarrierTest.cc' in bld(use=['ns3-lte'], features=['cxx', 'cxxprogram'], idx=19, meths=['process_rule', 'process_source', 'apply_link', 'process_use', 'propagate_uselib_vars', 'apply_incpaths', 'create_task_macapp', 'create_task_macplist', 'set_macosx_deployment_target'], prec=defaultdict(<type 'list'>, {}), _name='lena-componentcarrierTest', posted=True, source='lena-componentcarrierTest.cc', ns3_module_dependencies=['ns3-lte'], mappings={}, path=/home/ns3dev/repos/ns-3-allinone/lenaCA/src/lte/examples, includes='#', is_ns3_program=True, target='ns3-dev-lena-componentcarrierTest-debug') in /home/ns3dev/repos/ns-3-allinone/lenaCA/src/lte/examples
=============================================================
On using find on the lte-ca dist, the file is not found ...

1. Any suggestions on where to get it from / replacing it /ignoring it / are welcome... please help
2. I will submit the CA code, as soon as, I can modify the gsoc code from where it was left off...
3. I also would like to know any particular tests you would like as proof for the CA functionality, using the updated code?
4. Which particular scheduler would you like to be modified for CA, any preference pl?
5. What additions would you like to be made in the stats module, particularly to show the relevance of increased bandwidth / RBs with
CA functionality added?

With Thanks and Regards,
madan

I will do it just for the sake of the challenge involved, and hope it will help, a lot of guys I see trying to do projects related 5G with
NS-3 but without CA or SDN or Cloud based nodes... One can push the envelop though gradually...

[Madan Pande]

 Marco,
                 Please ignore my earlier post today...

                I could compile the code successfully making a minor change or two? Though normal compilation after merging lte-ca with Lena code reports a missing file "lena-componentcarrierTest.cc", it may not be required at all for writing examples... and testing CA functionality,  If it is required to be created for Testing, please advise its required functionality ???

2. I will submit the CA code, as soon as, I have a working code with some examples...

3. I also would like to know any particular tests you would like as proof for the CA functionality, using the updated code?
4. Which particular scheduler would you like to be modified for CA, any preference pl?
5. What additions would you like to be made in the stats module, particularly to show the relevance of increased bandwidth / RBs with
CA functionality added?

I will do it just for the sake of the challenge involved, and hope it will help a lot of guys I see trying to do projects related to 5G with

NS-3 but without CA or SDN or Cloud based nodes... One can push the envelop though gradually...

With Thanks and Regards,
madan

P.S. I saw the route CTTC is taking with its test bed, hopefully it will add a lot of clarity in defining what could be 5G. right now
there are too many claims...

[Szabolcs Nováczki]

Hi Madan, Marco,

So you are working on the CA functionality? I have also started to work on the CA code (i have not modified anything jet, but trying to understand the code). So there are at least 3 parallel tracks. Don't you think it would make sense to join the forces? Please let me know!

BR,

Szabolcs

[Madan Pande]

Hi,
         In principle, such approaches are possible, though I am not clear on what your interest are...The code is quite buggy to begin with... then comes the issue of what you want as functionality, what I am working towards and what is Piotr's interest... Even if there is a common goal, without okey by NS-3 maintainers and reviewers,  it will end up as many unmaintained archives of NS-3...
        The maintainers only look after the code merged in to NS-3, and therefore they take lead, decide course to take ... often, these directions form the targets for the GSOC project are undertaken every summer ...Dizhi Zhou's work on Schedulers done under GSOC is an outstanding example of GSOC contribution... I donot think, it is possible to track GSOC targets with what some individuals may like to do...So my suggestion to you is, go on with your study, dirty your hands (if you need to) with NS-3 WiFi and LTE, LAA/LBT codes and await the announcements for the GSOC CA project targets... Try and Get linked to that...

2. In itself CA in LTE is no big deal now, plenty of work has been published and atleast one Indian IT company and one university in USA and one in China, one in Australia, have their versions of NS-3 which are way way advanced but their code archives are not public... Their published work using their version of NS-3 speaks voulmes of their ingenuity and collective effort... If possible, get linked to them...( a IEEE Digital library search... could help)

3. Also, the directions of the NS-3 or CTTC Lena code could be different... Is their interest 5G? Is their interest Rel 13 dense Hetnets... or LAA/LBT with CA... this is not clear to me from what so far Marco or Dr. Baldo have said in the users groups...

This is my two bits...

Regards,
madan

[Lorenza Giupponi]

Dear Madan,

I work in CTTC together with Marco on projects related with LENA. I can then clarify what is our roadmap with respect to the LTE module in general.

As Marco already mentioned in some other post, we are now retaking the Carrier Aggregation work, we will fix it, improve it and complete it, in the context of the LTE module, and also we will merge it to the LTE-Wi-Fi module. We estimate though that we will not be able to release this work before autumn 2016.

If in the meanwhile any of you has time to devote and is able to progress with the CA implementation, we will be happy to agree and discuss on the different options, in order to join efforts, provide feedback and speed up the work at least of the LTE module.

Kind regards,

Lorenza

[Madan Pande]

Lorenza,

                  That is good to learn...
2. After fixing the LTE code, I will write say two examples, one to test the CA working and another to exemplify a typical HetNet scenario, ofcourse the current code does not cater to cross carrier scheduling which is a disappointment...if you so wish, may be we can discuss such changes, or what you want offline from this group...

With Thanks and Regards,
 madan

[Lorenza Giupponi]

Hi Madan,

thanks for your interest in contributing, let's stay in touch then.

About cross carrier scheduling, this is a long shot, and would radically change the design. In addition it is not a mandatory functionality. We are still not very much convinced that it would be worth the effort.

Our idea is to start by gradually increasing the interactions between the two CC schedulers. We believe that whatever add in this direction would certainly improve the CA functionality. But we can certainly discuss in case you are interested in working on a cross-carrier scheduling function/architecture. Your help would be more than welcome.

Lorenza

[Piotr Lechowicz]

Hi,

If that would be of any help I ported the code to 3.25 and removed quite a few bugs to make work for 1 CC, at least. The result I stored at https://github.com/piotrlech/ns-3-LTE-Carrier-Aggregation. Currently it still crashes for 2 CC's because the 

LteUePhy::ReportRsReceivedPower (const SpectrumValue& power) receives the empty power from time to time.

BR,

You received this message because you are subscribed to a topic in the Google Groups "ns-3-users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/ns-3-users/zv46lm6RSK0/unsubscribe.
To unsubscribe from this group and all its topics, send an email to ns-3-users+...@googlegroups.com.

[Lorenza Giupponi]

Thanks a lot, Piotr!

Lorenza

[Manoj Mishra]

Hi  Piotr

I am also working on carrier aggregation for multiple carriers as my master's thesis.

May I know that what changes are needed to run for single carrier?

May I get your  emailID to diascuus more about CA as I want to write test case ?

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 https://groups.google.com/group/ns-3-users.
For more options, visit https://groups.google.com/d/optout.

--
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 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.

--
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 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.

--
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 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.

[Piotr]

Hi there,

Any progress with respect to the CA debugging?

BR,

[lorenza....@gmail.com]

Hi Piotr,

The latest development version is available here: http://lena.cttc.es/hg/lena/

This includes also some of the fixes you shared.

Lorenza

Inviato da iPhone

[Piotr Lechowicz]

Wow, chapeau bas!

Ms. Biljana must be a genius or a magician or both!

Thanks,

On Sat, Aug 20, 2016 at 11:40 AM, <lorenza....@gmail.com> wrote:

Hi Piotr,

The latest development version is available here: http://lena.cttc.es/hg/lena/

This includes also some of the fixes you shared.

Lorenza

Inviato da iPhone

Il giorno 19 ago 2016, alle ore 18:41, Piotr <piotr.l...@gmail.com> ha scritto:

Hi there,

Any progress with respect to the CA debugging?

BR,

On Thursday, April 7, 2016 at 1:32:45 PM UTC+2, Piotr wrote:

Hi,

Where the source code of the GSOC2015LTECA is stored? I have downloaded the code from http://code.nsnam.org/daniloa/ns-3-dev-lte-ca and it turned out to be quite buggy, I'm afraid.

BR,

--
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.

--

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 a topic in the Google Groups "ns-3-users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/ns-3-users/zv46lm6RSK0/unsubscribe.

To unsubscribe from this group and all its topics, send an email to ns-3-users+unsubscribe@googlegroups.com.

[Madan Pande]

Piotr,

         My two bits...

2.     The implementation is primarily for the NOOP case, you would need to extend or make a derived class from the ComponentCarrier management (atleast for the Enb one), to use two CC (the max currently aimed at). Also, do likewise for the UE if you care to have more flexibility, and do a version of CC algorithm from the given one... Touch the SAP classes involved if needed...

                 That is how I had worked with it...

Cheers !

- madan

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

[Ramakrishnan]

Hi, 

It seems http://lena.cttc.cat/hg/lena/ isn't functional. Can you please point to any mirror site that hosts the carrier aggregation updates?

Thanks, 

Ramakrishnan

[Marco Miozzo]

Hi,

the CA code has been already integrated in the official release, you may find it in the ns-3 dev repository, http://code.nsnam.org/ns-3-dev.

best,

marco.

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

[Ramakrishnan]

Thanks, I am able to spot the changes. Couple of questions:

1) Is there an example code anyone has written that builds the e-2-e setup and uses multiple CC. (like lena-simple-epc.cc but with multiple CC usage)?

2) Is there any issue if number of CC is increased from 2 to 3 or 4?

Thanks, 

Ramakrishnan

[Biljana Bojovic]

Hi Ramakrishnan,

the lena repository is currently under migration process so some services are not available. Sorry for the inconvenience.

As Marco said the CA code can be found at ns-3-dev as well the documentation.

1) For configuration of carrier aggregation please see the description here: https://www.nsnam.org/docs/models/html/lte-user.html#carrier-aggregation-examples.

2) Yes, you may configure up to 5 CCs.

Cheers,

Biljana

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

[Ramakrishnan S]

Hello Biljana, 

Thanks, I will look through the configuration examples you have mentioned. 

Just few additional questions:

a) In the current implementation, is there a way to restrict a UE to a specific Carrier component - through Attach procedure or alternate Helper code? .e.g. eNodeB has CC1 and CC2 configured. Can UE1 be made to use "CC1 and CC2" for Uplane traffic but UE2 restricted to use only CC2 for Uplane (and not CC1).

b) If above restriction could be placed, Can the scheduling be done differently for UE1 and UE2 in CC2? 

c) Also, for any different allocation of packets to each component carriers (other than RR), should the changes be done in NoOpComponentCarrierManager/RrComponentCarrierManager ?

Thanks so much, 

Ramakrishnan

--

-----------------------------------------
S. Ramakrishnan