Re: [Discuss-gnuradio] Regarding the updating and extending gr-radar idea for GSoc 2018

[Martin Braun]

On 02/24/2018 05:34 AM, suraj hanchinal wrote:
> Hello Everyone,
> I am planning to apply for GSoc 2018. I was very interested and would
> love to work on the gr-radar toolbox. I was interested specifically in
> implementing passive-radar and multi-antenna support to gr-radar. Since
> a passive radar system would need a multi-antenna support anyways, I
> think they could be implemented together. I have been reading some
> research papers regarding passive radars to get a better understanding
> and I think implementing a passive radar to use FM transmitters nearby
> to locate objects as a good starting point and then turn to using wifi
> access points in an area as the next step. I am currently developing a
> proper write-up as well for the same. I would kindly request you to
> point out any things I should know of or just general suggestions on the
> subject. I would also kindly request the mentors Martin Braun and Stefan
> Wansch to enlighten me on this.

Suraj,

thanks for your interest in GSoC and gr-radar! Passive radar would
indeed be a pretty nifty addition. It's probably easier to get better
results using more high-bandwidth signals with better cross-correlation
properties, such as terrestrial TV stations.

Note that I assume that you have some familiarity with the signal
processing fundamentals of passive radar. That said, you can do a lot
better than cross-correlating when doing passive radar with OFDM
signals, but that's a little too much to lay out in a short email (in a
nutshell, you can recreate the original tx'd signal and use symbols as
samples in the time-frequency domain to generate range/Doppler diagrams
without having to worry about cross-correlation artefacts).

Also go check out Juha Vierinen's work (e.g., this:
https://hackaday.com/2015/06/05/building-your-own-sdr-based-passive-radar-on-a-shoestring/,
an old blog post I googled real quick). And of course Jean-Michel's work
that Phil mentioned.

-- M