[Discuss-gnuradio] Module for MAC layer development | GSOC'19 Proposal

Hi GNU Radio Community,

As suggested by Marcus, I have made a design plan which can be found in the attached proposal for GSOC.

The following is an introduction to the proposal.

GNU Radio gives an open platform for implementing software radios using signal blocks. Currently, GNU Radio's default blocks allow us to only build Physical Layer. Due to this restriction, not more than two devices can talk in the same wireless channel. If a sub-layer Medium Access Control (MAC) is implemented, this restriction can be resolved. 

GNU Radio provides a great opportunity for students to have hands-on experience in implementation. However, due to the restriction up to the physical layer, experimentation in MAC layers is not possible. This proposed module extends the experience to more areas in communication and networks and at the same time adds the features of random and controlled access.

My previous experience, projects I worked on and contact details can be found in my CV at the following link.

https://github.com/pramodhrachuri/My-Potatos/raw/master/CV_Pramodh.pdf

The latest iteration of this Proposal can be found at the following link.

https://github.com/pramodhrachuri/GSOC_Proposal/raw/master/GSOC'19_Proposal.pdf 

I can share more details that may help you take the decision if asked for.

Regards
Rachuri Sri Pramodh,
B. Tech (Hons) 3rd Year,
Department of Electrical Engineering and Computer Sciences,
IIT Bhilai  

On Mon, Mar 11, 2019 at 3:49 PM Müller, Marcus (CEL) <mueller@kit.edu> wrote:

Hi Rachuri,

thanks for reaching out!
So, you're planning on applying for with own project idea – which is
great; it shows enthusiasm and a high confidence you know what you're
signing up for. Problem is: while for all the ideas on the ideas list
mentors are already assigned, someone from the GNU Radio developer
community would need to be willing to mentor your specific project. At
this point, we don't have someone like that, but I'd assume most of us
haven't even read your mail yet. Still, this increases your risk of not
being accepted if we simply don't find a mentor.

So, because we need to find someone who'd be willing to mentor you,
you'd need to come up with a pretty detailed design plan, so that we
get the picture of what you want to implement. This would especially
mean you'd have to address the foreseeable problems in your proposal.

From the top of my head, these would be dominated by timing
uncertainty, and how to modularly implement things so that e.g. the MAC
mechanism or the flow control can be exchanged. Timing uncertainty is
what usually comes hardest for SDR MACs – you'd need to address how
you'd solve the problem that the delay (between receiving some data to
having detected / calculated a time at which you'd want to send ) might
be pretty random, and also pretty large.

All in all, this is a very open-ended problem, and to me, it sounds
very much like a systems architecture problem that needs to come up
with a clean and useful API that has both some educational value as
well as actually working over the air. A large project, even for
experienced developers!

Best regards,
Marcus

On Sun, 2019-03-10 at 00:43 +0530, Rachuri Sri Pramodh wrote:
> Hi GNU Radio Community!
> I am a pre-final year student in Dept of EECS at IIT Bhilai.
> We were thought about how to use SDRs using GNU Radio during one of our lab courses and also made a small python-code block as an assignment. It was then when I got the idea of making a block for a Datalink layer implementation.
>
> The block would have options to configure Framing, MAC, Addressing, Flow and Error control. If this block is properly implemented, one should be able to develop a small wireless network of multiple SDRs. This block will help students who study Wireless Networks in experimenting and applying various MAC models and even make some of their own.
>
> Later, when I saw GNU Radio among the list of GSoC's participating companies, I got really excited and wrote this mail. Please let me know your thoughts on this proposal for my GSoC's entry.
> I wrote a very brief description of my proposal here and I am ready to submit a very detailed proposal soon if asked for.
>
> Regards
> Rachuri Sri Pramodh,
> B. Tech 3rd Year,
> Department of Electrical Engineering and Computer Sciences,
> IIT Bhilai
>
> _______________________________________________
> Discuss-gnuradio mailing list
> Discuss-gnuradio@gnu.org
> https://lists.gnu.org/mailman/listinfo/discuss-gnuradio