Re: [Discuss-gnuradio] [GSoC2018] Adding Passive radar and multiple device support to gr-radar toolbox

[Martin Braun]

On 03/23/2018 11:26 AM, Müller, Marcus (CEL) wrote:
> Hi Suraj,
> 
> thank you very much for sending us your proposal! This already looks 
> very nice. Can I wish for you to add something like a rough block
> diagram that describes how your simulated illuminator, your signal
> recoverer, your radar estimator and your clutter reducer work
> together, and which information you plan on letting these exchange?
> That would allow us to "mentally" map what you're doing each week in
> your proposed timeline to components of the system.
> 
> All in all, this is pretty ambitious, but exciting!


Sure is!

There's a couple of typos in the proposal, and I always think that
distracts from the quality of content and the content itself.

My main issue with this proposal is its vagueness. I would recommend you
expand the sections that explain the algorithms you want to implement *a
lot*. In particular, we need to know if you've understood the underlying
math, DSP, and implementation requirements.
Do you have thoughts on a clutter removal algorithm? Like Marcus says,
it will better to do a simple approach first. You could, e.g. remove all
zero-Doppler targets.
As a result of this vagueness, I believe the timeline is probably
inaccurate.

GUIs are nice, but you should either make them stretch goals, or actual
first-class citizens.
It's always better to limit your main deliverables to make sure there's
time for cleanup and merging. What if the gr-radar maintainer has issues
with your pull requests? You're planning to add a of stuff. Were you
planning on submitting PRs continuously? If so, I recommend writing that.

I'm also curious how you do your signal recovery.

The overall proposal is interesting, though, no doubt about that!

-- M





> How will you tackle the OFDM signal recovery? I think your reference 
> [2] is really much to be completely done in one GSoC, so it would be 
> totally OK to say you just picked a reduced approach. Still, if you 
> want to do that in all its glory, that would be cool, too, but I'd
> ask Martin how much work he'd expect that to be, and if necessary,
> reserve more time for the algorithmic part alone. I'm also including
> Jean- Michel Friedt of low-cost passive radar fame[A], as I hope he
> might have a moment to read and comment on your proposal.
> 
> Best regards, Marcus
> 
> [A] http://jmfriedt.free.fr/URSI.pdf On Fri, 2018-03-23 at 12:54
> +0530, suraj hanchinal wrote:
>> Hello Everyone, I am Suraj Hanchinal, a second year undergraduate
>> in Electrical Engineering at the Indian Institute of Technology,
>> Kanpur. I had approached the mailing list and communicated with
>> Martin Braun, the mentor and others regarding the gr-radar toolbox
>> extension idea. I decided to work on adding passive radar support
>> to the toolbox after these discussions. I have finally completed
>> the proposal [1] and I would like feedback as well as suggestions
>> for improvement on the proposal.
>> 
>> Thank you,
>> 
>> Regards, Suraj Hanchinal
>> 
>> Proposal [1]
>> https://github.com/surajhanchinal/GSoC_proposal/blob/master/My%20GSoc%20Proposal.pdf
>> 
>> 
>> _______________________________________________ Discuss-gnuradio
>> mailing list address@hidden 
>> https://lists.gnu.org/mailman/listinfo/discuss-gnuradio
>> 
>> 
>> _______________________________________________ Discuss-gnuradio
>> mailing list address@hidden 
>> https://lists.gnu.org/mailman/listinfo/discuss-gnuradio