Thanks Dan! It was great seeing you at GRCON, and I learned a lot of DSP and Python from taking your classes in 2020. Thanks! For everyone else, please check them out here: www.dsprelated.com/courses
Another excellent video with solid resources. Really appreciate sample sample data sets being made available for people without hardware access to learn and experiment. I do hope you continue to add to this series as time and topics permit.
I'm glad to see another video!! I haven't seen anything like your radar series anywhere else, thank you so much for putting together this material and helping so many people understand this technology better 💪🏻
Good to see you back. Not sure if you have covered this, but have you considered applying cross correlation to dig out the signal from under the noise? Once you have a usable return, you could then apply a Kalman filter to track the drone.
Hello Jon, Great Video as usual, I just wrote to the Antenna Supplier to arrange shipping because they dont ship in canada at the moment. May I ask if you could provide us, your followers, with a discount code for the CN056 to make it a little more affordable ?
The tests show that Doppler radar returns for a stationary target or target moving perpendicular to the radar illumination axis do not generate significant returns. Given the subtractive corrolated chirp suppressing the background returns I'm wondering if a diffeerntial range target tracking vs background range signaling could handle those cases.
@@jonkraft Unfortunately no real ability to do so and . The form you use is essentially a target on a cross vertical plane. I guess what I'm proposing is essentially creating a 2d horizontal planar correlation to highlight targets in a positional map using sort of a synthetic dopler. If multiple mapping planes were generated, then an stationary ariel target is above ground level and should still show against the suppressed background... maybe.
The phase shift between adjacent chirp pulses carries the Doppler information. Will it get affected if we apply phase alignment in the 2 pulse canceller?
It's a good thought, but no, the velocity is not affected by the 2 pulse canceller. You can see that in the plots where I show both 2 pulse and no pulse (regular). The velocity stays the same. The velocity is the change in distance (beat freq) from one pulse to the next.
That is a really good question. Unfortunately, I don't have any RF models for the Phaser. There is a Keysight SystemVue model for the ADAR1000. But there really should be an RF model for the whole thing....
Details on the phaser kit, included price and availability are here: www.analog.com/cn0566 There's details on size and construction on the wiki page. But it's not intended to be used in any production. It is a learning/exploration platform for phased array radar concepts and prototyping.
I hope so! But if people do implement other algorithms with that radar data, then I will do a video on their contributions. That would be cool to learn. But otherwise, I'm moving on to some things with Pluto and phased array beamforming.
Huh. "Two Pulse" looks like a high pass spacial filter. It is the derivative (difference) in image processing or image reconstruction from projections, etc. I must make this! Maybe lower frequency with a HackRF.
@@jonkraft I wonder if you did not do the two pulse processing and just did high pass on the output image. A simple Gaussian hi-pass kernel. Would it be the same? Oh, wait. The data is available :-) (If it worked you could do it on a graphics card easily)
