Re: [Discuss-gnuradio] Introduction for GSoC participation

[swapnil negi]

Hello everyone!

Happy holi! (holi is an Indian festival of colors)

I am done with the tutorial on out of tree modules. I experimented with the same and am amazed with the beauty of gr-modtool.

But while running make test, I found out that 'make test' gave two different results on the same test files and C++ files, i.e. I ran the same command twice and got two different outputs. And sir if you can please guide me with what to do next, I will be highly thankful.

PS: Thanks for such informative responses. They were highly beneficial.

On Mon, Feb 26, 2018 at 9:21 PM, Nicolas Cuervo <address@hidden> wrote:

Hi there!

We are very glad that you find interest in contributing to GNU Radio and GSoC! You are definitely on the right road by getting familiarized with the software and its perks by installing it and following the guided tutorials. If you have any question regarding functionality that you want to clear out before starting writing a proposal, please do not hesitate on letting the mailing list know, and we certainly will help you around it.

Now, the ideas that you chose are tasks that are essential behind the scenes of GNU Radio's functionality and improvement on any of them will be immensely appreciated by the whole community. As you noticed, proficiency in both c++ and python is key to getting the ctrl port and the gr-modtool to the next level. There are a couple of weeks before the application date, so I would suggest you to start choosing which task seems more interesting for you and give it a bit more focus on getting hands-on experience. You can post your progress on this thread, or ask the mentors (me included) about aspects that help you refine the scope of your proposal. 

As I'm listed as mentor for the gr-modtool idea, here a bit more about it: you mention the if-then-else rules in some of the options of the tool, and definitely, that is something that needs work on. However, the intention of this as a GSoC project goes beyond that and here, given the whole lot of tasks that are part of the "wish-list" of the tool, a strictly definite scope is a *must* for a student that wants to take this task. As the idea title says, we are facing a "complete makeover". But fear not! you will be not needed to make miracles in record time, and you won't be alone! The first step in this process is knowing the tool, so go ahead and create a couple of dummy blocks as per [1], check the code to understand what is going on in the background and determine if you see the stuff that can be done better. Regarding the bullet that says "rewrite as a plugin architecture", you can also have a quick overview how rfnocmodtool [2] and volk_modtool [3] work, just to have a feeling of what kind of plugins are expected in the future (writing the rfnocmodtool and volk_modtool plugins is not part of the project idea, by the way). And again, questions are very welcome!

+Marcus might have some short insights on the CtrlPort project that might help you make your choice.

Cheers!

- Nicolas

[1] https://wiki.gnuradio.org/index.php/OutOfTreeModules

[2] https://kb.ettus.com/Getting_Started_with_RFNoC_Development#Starting_a_custom_RFNoC_block_using_RFNoC_Modtool

[3] http://manpages.ubuntu.com/manpages/artful/man1/volk_modtool.1.html

On Mon, Feb 26, 2018 at 3:53 PM, swapnil negi <address@hidden> wrote:

Hi,

I am Swapnil Negi, an Electronics and Communication undergraduate at Indian Institute of Technology Roorkee, India. I am highly interested in contributing towards GNU Radio as my GSoC project. 

I am done with setting up the project and have gone through the guided tutorial on gnuradio-companion. Right now, I am learning to program GNU Radio in python and C++.

I am a member Information Management Group, a student group responsible for maintaining institute's website and intranet applications; Programming and Algorithms Group, a student group aimed at spreading culture for programming and mathematical algorithms among students. So, I am moderately fluent in python and C++.

I have had courses on "Signals and Systems" and "Principles of digital communication" in my academic curriculum which I found pretty interesting, so I am highly enthusiastic about working with GNU Radio this summer.

I have checked out the ideas list. Two of the ideas suit me:

1. CtrlPort backend implementation: I, myself faced issues with the thrift version so I really wish to improvise this. I have started reading about remote procedural calls, message queues, etc, the differences and benefits of different message queues like level of abstraction, ease of implementation, etc. It seems interesting to me.

2. gr-modtool overhaul: I haven't gone through the code structure of gr-modtool but the concept is really interesting. It will also help me get the real feeling of GNU Radio. I saw the present series of if's and else's and would like to work on improvising this. 

I am not very sure of how to start contributing to the project, so it will very helpful if I can get your guidance. I will be highly obliged.

Thanks.

_______________________________________________
Discuss-gnuradio mailing list
address@hidden
https://lists.gnu.org/mailman/listinfo/discuss-gnuradio