Hi Ziad,
To give you some feedback, I'll go out of order:
3) I believe deciding on the exact library to use is an important first step, and is part of the work to be done as it requires time to try out the different options. So having that decision as part of the gsoc itself is a good thing. It could even be thought of as a deliverable
1) The discussion of pros and cons is good, but the exact choice might be dependent on the specificities of the chosen library. For instance, ImGUI might be easier to integrate directly into C++ blocks than Bokeh's Python
2) In general, it's good practice to test as you go. Making all the sinks and widgets before starting to test them might leave you with nothing working, or having to rewrite everything if you find systemic bugs in your code or architecture.
Expanding on the topic of risks. It could be good to think about, and detail where you could have some wriggle room to tackle unexpected issues and/or plan B, prioritization in case time runs short.
Cyrille MORIN Ingénieur SED
Équipe MARACAS
Centre Inria de Lyon
Laboratoire CITI
Campus La Doua - Villeurbanne
6 avenue des Arts
F-69621 Villeurbanne
Le 27/03/2026 à 18:52, Ziad Haithem 202201027 a écrit :
Sorry for the delay in my reply, things took a bit longer than i expected
Thanks for the answers
I went ahead and looked up some developer tutorials for GR4 and i think i have a good enough of an understanding now to be able to write a good proposal,
Here is the google doc fo my proposal with the technical aspects complete and that's mainly what i would like to get feedback on
Specifically what i would like feedback on is the following
1) My main design choice on the core infrastructure and if my rationalization behind it is sufficient (I detail 3 approaches in my proposal and compare them)
2) I lack quite a bit of experience with QA & documentation and so those parts of the proposal may be a bit lackluster any feedback on that would definitely help
3) I have essentially pushed back the question of what library for browser plotting will be used (Bokeh, imgui, etc) to being part of the first week of work, as i feel i wont have enough time to do an actual good assessment before the proposal deadline, is that okay or do i need to specify the library which will be used ?
Obviously any feedback on anything else is welcome, I am sure I have so much to learn !
Since its a google doc, feedback via comments within the document would be easier than email i think, however whatever is preferable for you is fine
Hi Ziad,
Having a running GR 4.0 with a custom OOT is certainly a good start, well done!
I'm definitely not in the best position to give advice on GR4 code philosophy as I still need to start playing with it before GSoc starts.
But I believe you'll find the best and most up to date info in the gnuradio4 repo itself, including the readmes in the /docs folder, and also the 4.0.0-RC1 tag description which covers a lot of the current state
I mention Bokeh in the project description because it's the dependency we use for the current (GR3) version of the remote monitoring system, so we know it can cover the main use case.
But it is far from perfect so we are open to using an other library providing that same core use case of remote monitoring with plots and interaction with widgets, with the added bonus of not requiring anything installed on the remote computer.
So ImGUI, with its many flavours could be an interesting option with better performance, nice plots... as long as it shows an efficient way for remote operation outside of the local browser.
I hope it helps with some of your questions,
Best
Cyrille MORIN Ingénieur SED
Équipe MARACAS
Centre Inria de Lyon
Laboratoire CITI
Campus La Doua - Villeurbanne
6 avenue des Arts
F-69621 Villeurbanne
Le 24/03/2026 à 04:40, Ziad Haithem 202201027 a écrit :
Dear GNU Radio Community,
I have decided that my gsoc proposal/project will be on the "BokehGUI in GNU radio 4.0" idea
1. Proof of Concept & Progress
To get comfortable with the GR 4.0 , I've developed a small PoC . It uses a signal generator block into a custom OOT module that publishes data to a socket via ZMQ. A Python script then subscribes to that socket and plots the sine wave in the browser in real-time using Bokeh.
Main goal is to just show i was able to take something generated by GR4 and get it displayed in the browser
I've also worked through the GR3 beginner and OOT tutorials, and spent time playing with ZMQ/Bokeh to ensure I can handle plotting and the data flow between processes.
2. Seeking Guidance on GR 4.0
While I'm thrilled I got the PoC running, this is my first time working with a codebase of this scale. My main concern is moving from "making it work" to "making it right." What are some resources that'll help me design and write code that "fits" with the design philosophy behind OOT modules for GR4. I feel like this will be very important for my proposal
3. Use of bokeh
I wrote the POC in bokeh and focused on it when experimenting because its what was mentioned in the project description and there are plenty of resources for it. However while researching I found that in
European GNU Radio Days in the next generation remote GUI section. "
Imgui" was considered a good candidate to be used for wireless plotting for GNU Radio 4.0. Has the community moved away from this opinion ? is it still present ? Should I write my proposal with only one library in mind or be flexible and have this been decided later on in development ? 4. Why This Project?
I initially looked for something with a bit more DSP and communications concepts. However this project feels important, and I decided to go with something which would have an impact instead of wasting time thinking of a custom project to propose that had me playing with some of the concepts i was most interested in.
I am really looking forward to the possibility of contributing to GNU Radio this summer as part of GSOC. Thank you for your time and for any pointers you can provide!
cyberspectrum is best spectrum (I think the code word is for the proposal only but one can never be too safe lol)
Sincerely,
Ziad
-- Ziad Haithem Fahmy
ID:202201027
Communications and Information Engineering Student
