GSoC 2017 | Application for "Enabling LTE CA handover to secondary cell"

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

GSoC 2017 | Application for "Enabling LTE CA handover to secondary cell"

Alexander Krotov-2
Hello everyone,
I'm Alexander Krotov, PhD student at Moscow Institute of Physics
and Technology [1].  I want to participate in GSoC 2017.  The project
I chose is "Enabling LTE CA handover to secondary cell" [2].

Biljana Bojovic is listed as mentor for this project, so I added
email to CC.

I have sketched a plan for my work, please review it and provide
some feedback. Then I will convert it to PDF and submit to GSoC
form.

My plan is as follows.
I want to improve LTE CA model architecture in several steps:

1. Make sure "secondary" cells are treated equally with "primary"
cells at eNB. The distinction should remain at UE and at UE management
entities on eNB only at this point.

2. Replace several Spectrum channels created by LteHelper with one
Spectrum channel to make transition between cells easier.  There
should be no need to switch Spectrum channels.  PHY should be
reconfigured but not linked against different channel object.  Due
to patches [3][4] I have submitted for Spectrum channel before the
overhead should be minimal.

3. Improve CCM <-> MAC interface on eNB to make sure CCM is informed
about changes made for each UE configuration.  Add the ability to
force reconfiguration of UE without actual RRC negotiation. Write a
test that shows how to manually change secondary cell.

4. Refine RRCConnectionReconfiguration message handling to allow
reconfiguration of secondary cells in the middle of simulation with
proper RRC negotiation.  Test that UE is able to receive reconfiguration
message and reconfigure itself according to it.

5. Improve CCM <-> MAC interface on UE to make sure CCM has enough
information to make decision about reconfiguration. Implement some
example algorithm, such as round robin with a predefined time period
that will decide when to handover.  Write a test scenario or example
showing how to use it.

6. Implement some more realistic algorithm or, even better, integrate
existing event-based handover algorithms.

7. If everything is implemented and time is left, improve measurements
collection and add algorithms to showcase their usage.

Some words about my skills.  I have absolutely no problems working
with C++, Python, Mercurial, Git, gdb, valgrind, address sanitizer,
Linux and FreeBSD.  As for patch reqirement [5], I have contributed
63 commits [6] to ns-3 in core, network, wifi, spectrum and lte
modules.

I use NS-3 on day-to-day basis and have contributed most of my
bugfixes and improvements upstream, but had no time to ship major
improvements so far due to lack of time required to prepare them
for code review. I hope to use GSoC as an opportunity to dedicate
time specifically to write mergeable code and improve it
as required during code review.

[1] https://mipt.ru/en/
[2] https://www.nsnam.org/wiki/GSOC2017Projects#Enabling_LTE_CA_handover_to_secondary_cell
[3] https://github.com/nsnam/ns-3-dev-git/commit/d8509c1a5888b804acc6a37f0b99a63673642b10
[4] https://github.com/nsnam/ns-3-dev-git/commit/16a4d770ef9cbee821ec8ec95554116a862e84ec
[5] https://www.nsnam.org/wiki/GSOC2017PatchRequirement
[6] https://github.com/nsnam/ns-3-dev-git/commits?author=ilabdsf
Reply | Threaded
Open this post in threaded view
|

Re: GSoC 2017 | Application for "Enabling LTE CA handover to secondary cell"

Biljana Bojović
Hi Alexander,

I am glad that you are interested in this project. Your plan looks very
good, I think that you covered all main points. Point 2 is very good
remark. I would make some small modifications in step 5, instead of
creating round robin algorithm we could just use lteHelper's procedure
to trigger the handover. Anyway, this is a minor modification, and I
maybe you had something different in mind, so this should be left for later.

Since the deadline for submission is in few hours
(https://developers.google.com/open-source/gsoc/) I think that you
should go on with GSOC application using this plan. If project is
approved, I would like that we discuss in more details about the project.

Best regards,

Biljana


On 31/03/17 01:32, Alexander Krotov wrote:

> Hello everyone,
> I'm Alexander Krotov, PhD student at Moscow Institute of Physics
> and Technology [1].  I want to participate in GSoC 2017.  The project
> I chose is "Enabling LTE CA handover to secondary cell" [2].
>
> Biljana Bojovic is listed as mentor for this project, so I added
> email to CC.
>
> I have sketched a plan for my work, please review it and provide
> some feedback. Then I will convert it to PDF and submit to GSoC
> form.
>
> My plan is as follows.
> I want to improve LTE CA model architecture in several steps:
>
> 1. Make sure "secondary" cells are treated equally with "primary"
> cells at eNB. The distinction should remain at UE and at UE management
> entities on eNB only at this point.
>
> 2. Replace several Spectrum channels created by LteHelper with one
> Spectrum channel to make transition between cells easier.  There
> should be no need to switch Spectrum channels.  PHY should be
> reconfigured but not linked against different channel object.  Due
> to patches [3][4] I have submitted for Spectrum channel before the
> overhead should be minimal.
>
> 3. Improve CCM <-> MAC interface on eNB to make sure CCM is informed
> about changes made for each UE configuration.  Add the ability to
> force reconfiguration of UE without actual RRC negotiation. Write a
> test that shows how to manually change secondary cell.
>
> 4. Refine RRCConnectionReconfiguration message handling to allow
> reconfiguration of secondary cells in the middle of simulation with
> proper RRC negotiation.  Test that UE is able to receive reconfiguration
> message and reconfigure itself according to it.
>
> 5. Improve CCM <-> MAC interface on UE to make sure CCM has enough
> information to make decision about reconfiguration. Implement some
> example algorithm, such as round robin with a predefined time period
> that will decide when to handover.  Write a test scenario or example
> showing how to use it.
>
> 6. Implement some more realistic algorithm or, even better, integrate
> existing event-based handover algorithms.
>
> 7. If everything is implemented and time is left, improve measurements
> collection and add algorithms to showcase their usage.
>
> Some words about my skills.  I have absolutely no problems working
> with C++, Python, Mercurial, Git, gdb, valgrind, address sanitizer,
> Linux and FreeBSD.  As for patch reqirement [5], I have contributed
> 63 commits [6] to ns-3 in core, network, wifi, spectrum and lte
> modules.
>
> I use NS-3 on day-to-day basis and have contributed most of my
> bugfixes and improvements upstream, but had no time to ship major
> improvements so far due to lack of time required to prepare them
> for code review. I hope to use GSoC as an opportunity to dedicate
> time specifically to write mergeable code and improve it
> as required during code review.
>
> [1] https://mipt.ru/en/
> [2] https://www.nsnam.org/wiki/GSOC2017Projects#Enabling_LTE_CA_handover_to_secondary_cell
> [3] https://github.com/nsnam/ns-3-dev-git/commit/d8509c1a5888b804acc6a37f0b99a63673642b10
> [4] https://github.com/nsnam/ns-3-dev-git/commit/16a4d770ef9cbee821ec8ec95554116a862e84ec
> [5] https://www.nsnam.org/wiki/GSOC2017PatchRequirement
> [6] https://github.com/nsnam/ns-3-dev-git/commits?author=ilabdsf

Reply | Threaded
Open this post in threaded view
|

Re: GSoC 2017 | Application for "Enabling LTE CA handover to secondary cell"

Alexander Krotov-2
> I would make some small modifications in step 5, instead of creating
round robin algorithm we could just use lteHelper's procedure to trigger
the handover.

Step 5 is specifically to test that CCM or some other internal entity can
make decisions about handover. lteHelper should be possible to trigger the
handover since the end of Step 4.

> Anyway, this is a minor modification, and I maybe you had something
different in mind, so this should be left for later.

OK, I uploaded the same version converted to PDF.

On Mon, Apr 3, 2017 at 3:52 PM, Biljana Bojovic <[hidden email]> wrote:

> Hi Alexander,
>
> I am glad that you are interested in this project. Your plan looks very
> good, I think that you covered all main points. Point 2 is very good
> remark. I would make some small modifications in step 5, instead of
> creating round robin algorithm we could just use lteHelper's procedure to
> trigger the handover. Anyway, this is a minor modification, and I maybe you
> had something different in mind, so this should be left for later.
>
> Since the deadline for submission is in few hours (
> https://developers.google.com/open-source/gsoc/) I think that you should
> go on with GSOC application using this plan. If project is approved, I
> would like that we discuss in more details about the project.
>
> Best regards,
>
> Biljana
>
>
>
> On 31/03/17 01:32, Alexander Krotov wrote:
>
>> Hello everyone,
>> I'm Alexander Krotov, PhD student at Moscow Institute of Physics
>> and Technology [1].  I want to participate in GSoC 2017.  The project
>> I chose is "Enabling LTE CA handover to secondary cell" [2].
>>
>> Biljana Bojovic is listed as mentor for this project, so I added
>> email to CC.
>>
>> I have sketched a plan for my work, please review it and provide
>> some feedback. Then I will convert it to PDF and submit to GSoC
>> form.
>>
>> My plan is as follows.
>> I want to improve LTE CA model architecture in several steps:
>>
>> 1. Make sure "secondary" cells are treated equally with "primary"
>> cells at eNB. The distinction should remain at UE and at UE management
>> entities on eNB only at this point.
>>
>> 2. Replace several Spectrum channels created by LteHelper with one
>> Spectrum channel to make transition between cells easier.  There
>> should be no need to switch Spectrum channels.  PHY should be
>> reconfigured but not linked against different channel object.  Due
>> to patches [3][4] I have submitted for Spectrum channel before the
>> overhead should be minimal.
>>
>> 3. Improve CCM <-> MAC interface on eNB to make sure CCM is informed
>> about changes made for each UE configuration.  Add the ability to
>> force reconfiguration of UE without actual RRC negotiation. Write a
>> test that shows how to manually change secondary cell.
>>
>> 4. Refine RRCConnectionReconfiguration message handling to allow
>> reconfiguration of secondary cells in the middle of simulation with
>> proper RRC negotiation.  Test that UE is able to receive reconfiguration
>> message and reconfigure itself according to it.
>>
>> 5. Improve CCM <-> MAC interface on UE to make sure CCM has enough
>> information to make decision about reconfiguration. Implement some
>> example algorithm, such as round robin with a predefined time period
>> that will decide when to handover.  Write a test scenario or example
>> showing how to use it.
>>
>> 6. Implement some more realistic algorithm or, even better, integrate
>> existing event-based handover algorithms.
>>
>> 7. If everything is implemented and time is left, improve measurements
>> collection and add algorithms to showcase their usage.
>>
>> Some words about my skills.  I have absolutely no problems working
>> with C++, Python, Mercurial, Git, gdb, valgrind, address sanitizer,
>> Linux and FreeBSD.  As for patch reqirement [5], I have contributed
>> 63 commits [6] to ns-3 in core, network, wifi, spectrum and lte
>> modules.
>>
>> I use NS-3 on day-to-day basis and have contributed most of my
>> bugfixes and improvements upstream, but had no time to ship major
>> improvements so far due to lack of time required to prepare them
>> for code review. I hope to use GSoC as an opportunity to dedicate
>> time specifically to write mergeable code and improve it
>> as required during code review.
>>
>> [1] https://mipt.ru/en/
>> [2] https://www.nsnam.org/wiki/GSOC2017Projects#Enabling_LTE_CA_
>> handover_to_secondary_cell
>> [3] https://github.com/nsnam/ns-3-dev-git/commit/d8509c1a5888b80
>> 4acc6a37f0b99a63673642b10
>> [4] https://github.com/nsnam/ns-3-dev-git/commit/16a4d770ef9cbee
>> 821ec8ec95554116a862e84ec
>> [5] https://www.nsnam.org/wiki/GSOC2017PatchRequirement
>> [6] https://github.com/nsnam/ns-3-dev-git/commits?author=ilabdsf
>>
>
>