• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 32 subscribers
  • Views 3420 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.

TIDEP-0094: tidud93 Reference Guide<<TI Designs: TIDEP-0094：80-m Range Object Detection With mmWave Sensor Reference Design>>

line
Prodigy 40 points
Part Number: TIDEP-0094
Other Parts Discussed in Thread: IWR1642

in <<80-m Range Object Detection With IWR1642 mmWave Sensor Reference Design>> Document，As described in Table 1 of Section 1.3:“This specification will be improved over time by showing how higher-level algorithms can extend the maximum measurable velocity beyond this limit"，and Described on page 6："Through high-level algorithms, the maximum unambiguous velocity that can be detected is 90 kmph."
Can you tell us what is this algorithm, how to implement, thank you!

  • 0
    TI__Mastermind 30405 points

    Hello,

    Radial velocity is directly measured in the low-level processing chain as phase shift of the signal across chirps within one frame using a two dimensional FFT on the frame data. The maximum unambiguous velocity observable is then determined by the chirp repetition time within one frame. Thus higher Vmax requires closely spaced chirps. The velocity measurement is unambiguous only if  | ω | < 180o (i.e. π radians) as shown below.

    (Please refer to the mmwave-training-series training videos for basics of FMCW radar. Velocity estimation is explained in https://training.ti.com/intro-mmwave-sensing-fmcw-radars-module-3-velocity-estimation?cu=1128486) 

    There are algorithms  which can be run in higher layer processing (tracking etc) that can undo the ambiguity introduced due to phase rollover (| ω | > 180o) and can extend the maximum measurable velocity by a few times (3x, 4x or higher) the native velocity measured by the sensor.

    There are various approaches to doing this (Velocity extension using tracking, Velocity extension using multiple chirp profiles etc) and you can search the web for details on these. We don't have a Vmax extension algorithm implemented at this time but we are working on this and it will be released on TI.com once it is implemented.

    Regards

    -Nitin

  • 0 in reply to Nitin Sakhuja
    Prodigy 40 points
    Dear Nitin：
    Earlier, we thought that the speed extension algorithm was low-level algorithm (similar to the Doppler FFT signal processing algorithm), in fact, it uses a high-level algorithm (such as tracking data processing algorithm) to implement.Thanks！
  • 0 in reply to line
    TI__Mastermind 30405 points

    Hi Line,

    As mentioned in my post, there are various ways to extend the maximum measurable velocity beyond the native measurement from the sensor, and tracking is one of them.

    There are other techniques which are not based on tracking. One such way is to use multiple chirp profiles (with different start frequencies or different start times) and solve the resulting velocity equations to compute the phase roll-over count. Again, you should be able to find literature about these methods on the web. But yes, velocity extension is done after the basic FMCW FFT processing (range, velocity etc).

    Can we close this thread if your question was answered?

    Thanks

    -Nitin

  • 0 in reply to Nitin Sakhuja
    Prodigy 40 points
    Dear Nitin,
    This resolved my issue, you can close this thread.
    Thanks