• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 29 subscribers
  • Views 798 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Original question:

IWR1443: raw data

Linux/IWR1443BOOST: Conversion Between bins to meter and m/s; Conversion between antenna to degrees

Daniel L.
Prodigy 140 points
Part Number: IWR1443BOOST

Tool/software: Linux

Hi,

I am working with IWR1443Boost, ES3.0, SDK 2.1.

I have some doubts regarding the conversion of the bins into meters(range bin), and meters per second (for doppler bin). Also, I have a doubt regarding the virtual antennas index conversion into degrees. 

1) I am able to acquire the Log magnitude range which in each cycle comes as a vector with a number K of columns according to the number of range bins. e.g., if range bins = 128 -> K = 128. As I plot the data on one cycle is easily perceptible that the magnitude follows the target according to target distance, as I can see going back and forth on the plot as I move the target closer or further. My initial assumption was that each K bin could be converted to meter by doing : Range[meters] = Range_resolution*K. However, the results given by this conversion do not follow the value in reality, and there is a relatively high discrepancy to the correct values. Therefore, my first question is: Is this conversion that I am doing correctly? If yes, is it common to have a discrepancy as high as 0.2 meters?

2) For the doppler matrix I also have one doubt. Considering that there are 16 doppler bins, is the conversion between bins to velocity as simple as multiplying the doppler bin index with the velocity resolution? Furthermore, if there are 16 doppler bins which I assume half is for positive velocity and half is for negative velocity, where would be the '0' velocity point? I mean, is the |maximum radial velocity| greater or smaller than the |minimum radial velocity|?

3) Regarding the azimuth heat map. Considering that I have the azimuth heat map with 8 virtual antennas, and azimuth resolution is equal to 15 degrees. Can I assume that each antenna corresponds to 15 degrees giving in total 120 (+-60) degrees azimuth ?

Best Regards,

Daniel L.

  • Former Member
    0 Former Member

    Hello,

    For 1) and 2) you should be multiplying the bin index by the [range or doppler]_bin_resolution which is different then the resolutions defined by the chirp specification.

    •   Range resolution per range FFT bin = range_resolution*(num_samples_per_chirp/num_range_fft_bins)
    •   Velocity resolution per Doppler FFT bin = vel_resolution*(num_chirp_loops/num_doppler_fft_bins)

    To understand how the data is parse for the plots I would refer you to check the source code of the demo visualizer. Please refer to this thread to do so: 

    Amanda

     i- Range resolution per range FFT bin:                            (20)

    - Velocity resolution per Doppler FFT bin:                                                                           

  • 0 in reply to Former Member
    Prodigy 140 points

    Hi Amanda,

    Thank you for your reply. It has solved at good extent my doubts.

    Would you know about this part of the question as well "Furthermore, if there are 16 doppler bins which I assume half is for positive velocity and half is for negative velocity, where would be the '0' velocity point?" By what I have observed it seems like the doppler side that corresponds to positive velocity has one less bin than the side representing the negative side because when a target has zero velocity it remains on a bin that is on the half side of the positive velocity. However I am not sure if this is correct, would you know that?

    Also, would you be able to answer number 3 as well?

    Cheers,

    Daniel

  • Former Member
    0 Former Member in reply to Daniel L.

    Hello Daniel,

    Regarding Q2) Yes you need to circle shift the FFT bins. For your example of 16 bins: bins 1-7 are positive, bins 8-15 correlate to negative and bin 16 would be 0. Please refer to the GUI source code I previously mentioned if you need more details on the parsing and extraction of doppler. Please also note that in the newer SDK 3.x+ the device reports doppler as m/s instead of index.

    Q3) Regarding azimuth heat map please review the concept of angular resolution and antenna field of view. I am not sure I understand this question as it relates to azimuth heat map. Please see this thread that reviews the concept of angular resolution: 

    Amanda

  • 0 in reply to Former Member
    Prodigy 140 points

    Thank you very much!