• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 31 subscribers
  • Views 465 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.

IWR6843AOPEVM: Calculation of chirp parameters

Morten Nissov
Intellectual 380 points
Part Number: IWR6843AOPEVM

In the pursuit of min-maxing chirp settings I've been following the video series from The Engineer on the Go and the The fundamentals of millimeter wave radar sensors and Chirp Design documents regarding the calculations. I have some clarifying questions.

First, my general process is attempting to "lock" 2 attributes (range max, range res, velocity max, velocity res) and thereafter calculate the best possible values for the remaining parameters.

  1. Velocity resolution: a function of wavelength and frame time
    1. How is the frame time here effected with multiple transmitting antennas, assuming there's 3 Tx all of which transmitting separately. The total frame time should be T_c * N_frames per Tx. Is the velocity resolution resulting from T_c * N_frames or T_c * N_frames * N_tx?
  2. Radar cube calculation
    1. How can the radar cube memory size be calculated given N_tx, N_rx, N_range_fft, and N_doppler_fft. And does the angular resolution play into this as well?
  3. Getting minimum timing constraints
    1. max IF frequency: given by datasheet, e.g. in this case is 0.9 * 10MHz
    2. minimum idle time: guidelines from table 3 of https://www.ti.com/lit/an/swra553a/swra553a.pdf?ts=1668691585541
    3. ADC start time: are there any guidelines for this?
    4. ramp time: calculated from T_c, idle time, and ADC start time

I think I'll likely develop additional uncertainties throughout the process, but I figure answering these is a good start. I've tried to underline the main questions, but hopefully this overall though process makes sense.

  • 0
    TI__Genius 10730 points

    Hi Morten,

    Please see our sensing estimator tool. In this tool we have calculators so you can enter a velocity / range / range resolution / velocity resolution and design the rest of your inputs around that. 

    In the second tab of the tool you can see what the timing parameters may look like. You can also see the radar cube equation. 

    1. Velocity resolution: a function of wavelength and frame time
      1. Velocity resolution is resulting from T_c * N_frames * N_tx?
    2. Radar cube calculation
      1. Radar cube is Nfft range * Number of chirp loops * N_tx * N_rx
      2. Nfft range is POWER(2,CEILING(LOG(number of samples,2),1)), so usually it is the number of samples
    3. Getting minimum timing constraints
      1. Please see the tool linked above! It has timing error checking.

    Thanks,

    Angie

  • 0 in reply to Angie Mitchell
    Intellectual 380 points

    I have tried to use the sensing estimator, it has returned config files which do not work and crash the sensor for me previously, I made a different post about this. Aside from this it requires inputting values for all 4 of max range, range resolution, max doppler, and doppler resolution.

    In my application only max range and max doppler are constrained. So for me I think it is more convenient to lock the first two and then try to calculate a configuration which returns the bets values for resolution possible.


    With this in mind, is there any documentation regarding these minimum timing constraints? Also how can the velocity resolution be a function of N_tx? Each transmitted chirp, post FFT, covers the same 0->max velocity range independently right?

  • 0 in reply to Morten Nissov
    TI__Genius 10730 points

    Hi Morten,

    You are right, our sensing estimator is a helpful tool but it does not consider every single limitation of the processing chain as that can vary between different processing chains.

    The documentation and our tuning guides in the industrial toolbox (people counting specific) are the tools we have to get started.  

    The chirp repetition period varies with number of TX effecting maximum velocity and velocity resolution since maximum velocity and velocity resolution are inversely related to chirp repetition period.

    Thank you,

    Angie