GSoC 2014: Wireless Network in the loop

Dear All,

My name is Asma Afzal. I am currently pursuing a PhD in Electrical
Engineering (Wireless Communication ) from the University of Leeds, UK.

I am interested in working on the WiNeLo GSoC project. WiNeLo is an
in-the-loop simulation framework for communication networks based on the
GNU Radio software radio toolkit. gr-winelo mimics the behavior of
common
RF frontends such as the USRP, but instead of sending the signal over
the
air, a central server plays the role of the wireless communication
channel.
Arbitrary channel models can be simulated, by passing their respective
GNU
Radio processing block to the server. Since this whole setup is
completely
transparent to GNU Radio applications, it is at any moment possible to
switch between simulations and real-world tests.

After watching the FOSDEM video, I believe I can make (part of) the
following additions to the ongoing project

  1. Implementation of various analytical channel models on the server to
    mimic real wireless channel. (Rayleigh, Nakagami-m, Rician, Weibull
    fading,
    Lognormal Shadowing)

  2. Implementation of different transmission strategies such as CDM or
    interference free TDM with different digital modulation schemes.

A user will be provisioned to input various controlling parameters like
the
frequency, rate of symbol transmission, wireless environment
(indoor/outdoor), etc.

With reference to these points, I wanted to ask senior members of the
community regarding which tasks would be beneficial to the community
immediately?

I will appreciate your input to point out the exact goal.

I believe that my skills are particularly well suited for this project.

I did my Masters in Telecommunication and Networks from NUST, Pakistan.
I
studied Wireless communication and Wireless networks in detail. I am
familiar with modulation schemes and wireless channel impairments like
path
loss, slow/fast fading, shadowing, noise and interference. My current
research is based on stochastic modeling of wireless networks in
presence
of path loss and fading. I have recently presented a paper at Globecom
2013. The paper hasn’t come on IEEE Xplore yet but you can find an
electronic copy on this link.

http://ipt.seecs.nust.edu.pk/Pubs/Asma%20Globecom%202013.pdf

Although I don’t have prior experience with GNU Radios, but I have
strong
programming skills in C/C++ and Matlab. During my masters, I worked on a
collaborative project with Cypress Semi-Conductor, USA. The project
involved implementation of AES-128 encryption and a part of thermal
management system on Programmable System on Chip (PSoC). As a result of
successful implementation, I got two technical articles published
(listed
below).

http://www.embedded.com/design/safety-and-security/4417178/Enhancing-system-efficiency-of-embedded-encryption-applications

http://www.embedded.com/design/prototyping-and-development/4422955/Building-an-efficient--intelligent--and-flexible-thermal-management-system

I am a keen learner and I have always wanted to apply my theoretical
knowledge on practical systems. This is my chance to do that. It will be
extremely beneficial for my understanding of wireless systems and since
this project is so relevant to my research, I would like to pursue it
after
summer as well and contribute to the GNU Radio open source community.

Thank you.

Kind Regards,
Asma Afzal.

Hi Asma,

this sounds very interesting, especially the implementation of new
channel models would be a great enhancement to the WiNeLo project (and,
if implemented as standard GR blocks, to GR in general). Concerning this
matter - and as an additional option - one could possibly also improve
and extend the already existing gr-channelsounder OOT (see
https://github.com/gbaier/gr_channelsounder).

Your second thought is also a great idea, but to me it sounds more like
a demo application for the WiNeLo framework. I am really looking forward
to see such applications running on top of WiNeLo, but I think that we
should invest all the valuable development effort in the framework
itself or the hardware and channel models right now.

I am looking forward to read your opinion on these ideas!

Nico


Von: [email protected]main.invalid
[[email protected]main.invalid] im
Auftrag von Asma Afzal [[email protected]]
Gesendet: Montag, 10. Mrz 2014 10:57
An: [email protected]
Betreff: [Discuss-gnuradio] GSoC 2014: Wireless Network in the loop

Dear All,

My name is Asma Afzal. I am currently pursuing a PhD in Electrical
Engineering (Wireless Communication ) from the University of Leeds, UK.

I am interested in working on the WiNeLo GSoC project. WiNeLo is an
in-the-loop simulation framework for communication networks based on the
GNU Radio software radio toolkit. gr-winelo mimics the behavior of
common RF frontends such as the USRP, but instead of sending the signal
over the air, a central server plays the role of the wireless
communication channel. Arbitrary channel models can be simulated, by
passing their respective GNU Radio processing block to the server. Since
this whole setup is completely transparent to GNU Radio applications, it
is at any moment possible to switch between simulations and real-world
tests.

After watching the FOSDEM video, I believe I can make (part of) the
following additions to the ongoing project

  1. Implementation of various analytical channel models on the server to
    mimic real wireless channel. (Rayleigh, Nakagami-m, Rician, Weibull
    fading, Lognormal Shadowing)

  2. Implementation of different transmission strategies such as CDM or
    interference free TDM with different digital modulation schemes.

A user will be provisioned to input various controlling parameters like
the frequency, rate of symbol transmission, wireless environment
(indoor/outdoor), etc.

With reference to these points, I wanted to ask senior members of the
community regarding which tasks would be beneficial to the community
immediately?

I will appreciate your input to point out the exact goal.

I believe that my skills are particularly well suited for this project.

I did my Masters in Telecommunication and Networks from NUST, Pakistan.
I studied Wireless communication and Wireless networks in detail. I am
familiar with modulation schemes and wireless channel impairments like
path loss, slow/fast fading, shadowing, noise and interference. My
current research is based on stochastic modeling of wireless networks in
presence of path loss and fading. I have recently presented a paper at
Globecom 2013. The paper hasnt come on IEEE Xplore yet but you can find
an electronic copy on this link.

http://ipt.seecs.nust.edu.pk/Pubs/Asma%20Globecom%202013.pdf

Although I dont have prior experience with GNU Radios, but I have strong
programming skills in C/C++ and Matlab. During my masters, I worked on a
collaborative project with Cypress Semi-Conductor, USA. The project
involved implementation of AES-128 encryption and a part of thermal
management system on Programmable System on Chip (PSoC). As a result of
successful implementation, I got two technical articles published
(listed below).

http://www.embedded.com/design/safety-and-security/4417178/Enhancing-system-efficiency-of-embedded-encryption-applications

http://www.embedded.com/design/prototyping-and-development/4422955/Building-an-efficient--intelligent--and-flexible-thermal-management-system

I am a keen learner and I have always wanted to apply my theoretical
knowledge on practical systems. This is my chance to do that. It will be
extremely beneficial for my understanding of wireless systems and since
this project is so relevant to my research, I would like to pursue it
after summer as well and contribute to the GNU Radio open source
community.

Thank you.

Kind Regards,
Asma Afzal.

Dear Nico,

Thank-you for getting back to me. I understand the needs of the
community
for a GUI for existing functionality.

I am still interested in working on this angle of the project.

I took a look at Omnet++ and it is quite impressive. Its built on the
eclipse framework. And is also open source. An alternative possibility
could
be adding to the graphical side of the existing Gnu Radio Companion
(GRC).

“One aspect (and also a new feature) should be the visualization of the
network structure/layout and node distribution itself where one also
should
be able to randomly deploy nodes and thus create a network structure.”
Please correct me if I am wrong. The required functionality is a gui
like
Omnet++.
But with gr-winelo blocks representing nodes and channels. The user can
add/delete nodes. And select which channel blocks to put between them.

I would like to work on this project. However, for my proposal, I need
to be
a bit more specific regarding the project requirements so I can study
and
formulate my timeline/deliverables accordingly.

Could you please share any relevant resource on gr-winelo with me? apart
from the FOSDEM talk.

And also elaborate more on the requirements of the gui.
e.g. the underlying framework for gui? eclipse? gtk/qt? etc…

Kind Regards
Asma Afzal

/On Mon, Mar 17, 2014 at 5:33 PM, Otterbach, Nico
[email protected] wrote:
Asma,

First of all: I’m sorry for my late reply.

Last week I had a talk with some other GR developers and we came to the
conclusion that channel models alone are not of high impoortance right
now.
We did a brainstorming about the most import WiNeLo projects at this
time.
Hence, we all agreed on the fact, that we have to work on the framework
itself. As a consequence, I want to propose you another possible
project: We
need some kind of GUI for the already existing functionality. One aspect
(and also a new feature) should be the visualization of the network
structure/layout and node distribution itself, where one also should be
able
to randomly deploy nodes and thus create a network structure. There
should
something similar to the Oment+±GUI or something like that. I will also
update the GSOC/page during the day!

Are you still interessted in working on the Wireless Netowrks
In-the-Loop
project at this point? I really hope so, but I can understand if that
doesn’t fit to your plans anymore.

Best regards
Nico/


View this message in context:
http://gnuradio.4.n7.nabble.com/GSoC-2014-Wireless-Network-in-the-loop-tp46822p47026.html
Sent from the GnuRadio mailing list archive at Nabble.com.

Thank you for your response.

Considering the timeline of GSoC, I am more interested in implementing
channel models. That way I can stick to the software domain to test and
validate my work. Also, I have frequently used Rayleigh/Ricean/Nakagami
models in MATLAB and I am familiar with the math behind them.

The available GR channel model blocks in the following link use addition
of
sinusoids and filters to approximate a channel response.
http://gnuradio.org/doc/doxygen-3.7/page_channels.html

I found a more accurate implementation of Rayleigh and Rician fading on
GNU
radio on this link:
https://www.cgran.org/browser/projects/usrp_fading_simulator/trunk/src/lib?rev=68
It follows this paper:
http://volta.sdsu.edu/~amir/Alimohammad09aCompact.pdf

A number of other fading channels can be implemented as GR blocks.

  1. Log-normal shadowing can be implemented as standard GR block to model
    Large scale fading and it finds its use in a number of mobility related
    applications. I have found a paper relevant to its 2D simulation:
    http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=1247814

  2. Nakagami-m channel: Figure 3 of the following paper describes how
    correlated samples of a nakagami-m faded channel can be generated
    http://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=01412063

I couldn’t find a GR based implementation of a basic path loss model.
A simple block implementation for path loss is necessary- for various
environments with a distance parameter (source-destination) and a
parameter
of path loss exponent (range 2-5).

If the above mentioned channel impairments are successfully implemented,
they encapsulate Rayleigh, Rician, Suzuki, Composite Rician-Shadowing
environments.

Please give me your opinion about these. Is there any other specific
model i
should be focusing on? What will be a sufficient goal for my GSoC? I’m
not
sure if I am aiming too high at this point :).


View this message in context:
http://gnuradio.4.n7.nabble.com/GSoC-2014-Wireless-Network-in-the-loop-tp46822p46913.html
Sent from the GnuRadio mailing list archive at Nabble.com.

This forum is not affiliated to the Ruby language, Ruby on Rails framework, nor any Ruby applications discussed here.

| Privacy Policy | Terms of Service | Remote Ruby Jobs