I’m a beginner to this field, so sorry if my questions aren’t the best, or if I’m missing information.

Anyways, I’m working with raw data samples from a FMCW radar. Basically, a frame constitutes of [Rx_Ant, Chirps, Samples] in Python.

My first question is what exactly this shape means. For instance, what does it mean for frame[0][0][0] with value of like 0.400569 or something.

My next question would be how after applying fft, and giving that result to plot, how does it produce a range-doppler results of velocity and distance when the shape is as stated above?

Lastly, if I wanted to store these values in a data frame or dataset for as machine learning inputs, how should I store the values and what shape should I expect. Like should I store one value as a 2-D array with the shape [Chirps, Samples] or should I store the values as like one value contained an array of samples for one chirp?

Once again, Sorry if there’s not enough information or if the questions are poor. I’m just confused about these few points.

Thank you.

  • $\begingroup$ Do you have a text that you are learning from? Does it not answer at least some of these? Some of these, I think, are too basic for anybody to want to answer. For example, with the dimensions as you have them frame[0][0][0] if the first sample from the first chirp on the first receive antenna. That is just basic indexing and not even a radar question. $\endgroup$ Commented Feb 6 at 16:13
  • $\begingroup$ You should also split this into multiple questions. Processing every dimension of the data cube goes into different domains and considerations that must be taken. If you're truly a beginner, you have your work cut out for you as there's a lot of theory you need to learn, and this is assuming that software is not the problem. If you're asking how to interpret a three-dimensional array, then you have other basics to work on. $\endgroup$
    – Envidia
    Commented Feb 11 at 19:15


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.