Because of the holidays, TI E2E™ design support forum responses will be delayed from Dec. 25 through Jan. 2. Thank you for your patience.

  • State Resolved
  • Locked Locked
  • Replies 11 replies
  • Answers 2 answers
  • Subscribers 32 subscribers
  • Views 774 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

IWR6843: Max unambiguous range displayed in mmWave demoVisualizer

Kei
Intellectual 720 points
Part Number: IWR6843

Hello, TI team,

I have a question about 'Max Unambiguous Range' displayed in mmWave demoVisualizer.
Simply thinking, max unambiguous range is range resolution * sample per chirp I think.
But, in demoVisualizer, max unambiguous range is 0.8 * range resolution * sample per chirp.
Could you let me know the meaning of 0.8 ?

For example, when saple per chirp is 256, range resolution is 0.0436, max unambiguous range is
0.8 * 0.0436 * 256 = 8.929
please see the picture below.

Regards,
Kei

  • 0
    TI__Mastermind 32690 points

    The 0.8 is a safety factor to account for the Nyquist limit. Frequencies on the radar device correspond to distances in real life. So, if an object's distance is at the very edge of the Nyquist limit, then it may alias into a lower frequency bin if the object moves slightly. So, that safety factor is designed to protect against that potential ambiguity. The radar device should still be able to recover the full spectrum, but there may be ambiguities as you approach the Nyquist limit. The datasheet uses 0.8, which is why we have this factor, but it's possible to use 0.9 or even 0.95 depending on the risk tolerance of the project.

  • 0 in reply to Nathan Block
    Intellectual 720 points

    Hi, Nathan,

    Thank you for your answer.

    You don't mean that regarding distance from 'range max * 0.8' to range max, there is possibility of aliasing, correct?
    I suppose that low pass filter processing is included before range FFT, correct?
    So from 'range max * 0.8' to range max is ambiguous whether radar can detect objects or not. Because low pass filter can not decay sharply at Nyquist frequency.
    Is my reading correct?

    Regards,
    Kei

  • 0 in reply to Kei
    TI__Genius 17265 points

    Hi Kei,

    My apologies for the delay, Nathan is Out of Office today, but will get back to you tomorrow. Thank you for your patience.

    Best Regards,
    Alec

  • +1 in reply to Alec May
    TI__Mastermind 32690 points

    Hi Kei,

    Your reading is correct. This is to protect against aliasing.

    Best,

    Nate

  • 0 in reply to Nathan Block
    Intellectual 720 points

    Hi, Nathan,

    Thank you for your answer.

    Regards,
    Kei

  • 0 in reply to Nathan Block
    Prodigy 30 points

    Let me pose a question on the direction of the consequences. Doesn't this safety factor in practice decrease the effective bandwidth and therefore the range resolution?
    If I understood correctly, the LPF gives out an IFmax at 0.9*Fs and then comes the ADC sampling. Having a lower IFmax is like shortening the effective observation interval of the signal in the time domain, which is in the end what determines how much we can separate 2 targets in the frequency domain.

  • 0 in reply to Oscar
    TI__Mastermind 32690 points

    Hi Oscar,

    This does not refer to the Low Pass Filter. The filtering and sampling should be the same, it just refers to the maximum range that users can expect to get unambiguous results from.

    Best,

    Nate

  • 0 in reply to Nathan Block
    Prodigy 30 points

    Hi Nate, thanks for the reply.

    Ok understood, clear that they are different factors now. But then if we refer to the 0.9 factor that limits the IF BW depending on ADC choice, in point 2.1.1(https://www.ti.com/lit/an/swra553a/swra553a.pdf?ts=1665428379647), this also limits the max range. Is this as well included in the other factor you are talking about (0.8)?

    And coming back to my previous question, does the factor described in the SWRRA553 doc ultimately has an effect on the range resolution?

    Thank you.

  • 0 in reply to Oscar
    TI__Mastermind 32690 points

    Hi Oscar,

    Let me consult with the authors of the App Note. I will get back to you tomorrow.

    Best,

    Nate

  • 0 in reply to Nathan Block
    Prodigy 30 points

    Hi Nate, any updates? Thanks

  • 0 in reply to Oscar
    TI__Mastermind 32690 points

    Hi Oscar,

    The IF signal is still sampled at ADC sampling rate. Hence to compute the “maximum unambiguous range”  for the purpose of computing our range axis we do not need the factor of 0.9.

    An anti-aliasing filter is used before the ADC sampling to prevent aliasing from the objects greater than the max range. A safety factor of 0.9 is typically used to emphasize that a portion of the frequency spectrum (range-Axis) is very noisy because of the non-idealities of the anti-aliasing filter and the signals around DC (i.e. TX-RX coupling, front-end non-idealities) can also alias back to Nyquist Frequency. Hence we typically want to discard the data corresponding to 0.9-1.0x of the ADC sampling rate

    I don’t see how the anti-aliasing filter limits the bandwidth of the chirp. It limits the maximum frequency content that can be present in the IF signal (and hence the maximum range) but does not shorten the IF signal duration and hence would not change the bandwidth.

    Best

    Nate