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.

IWR1642: Start sensing by an external trigger signal

Part Number: IWR1642

Hello TI Expert,

I have got a few questions about IWR1642 from our customer. Would you please check these questions?

Q1: They want to input some trigger signal from externai signal and start sensing by arbitary timing.
Is it possible to do that?

Q2: If it is Yes in Q1, how long does it take from trigger to start sensing, to finish measurenent and to data transfer.
Would you please tell us each delay time ?
(1) Time From Trigger to Start Measuring
(2) Time From start Measuing to Stop Measuring
(3) Time From Stop Measuring to Data Transferring

Q3: If it is Yes in Q1, would you please tell us how to achieve this ? What signal and where can they input to?

Thank you very much for your comment.

Best Regards,
Nobu Arai

  • Hello Nobu,

    I've asked an IWR team member who has conducted tests on these to look at your question. He should have an answer for you sometime Friday.


    Cheers,
    Akash
  • Hello,
    Q1 - yes, the External Sync In - is a 3.3v IO signal that can be used to start the Frame, it must be selected in software.

    Q2 - (1) first portion, from External Sync Input to starting the Frame, is estimated to be < 5 usec, if the software has already been configured

    (2) this depends on the Frame, Profile, and Chirp design. The Sensing Estimator can help design these FMCW Radar settings
    Typically there are several programmed times given for each chirp, then the frame is a number of chirps, see SWRA553 application note.
    Frame Time Period = (numberChirps * Chirp TimePeriod) + Interframe timing
    Chirp TimePeriod = RampEndTime + IdleTime
    Depending on the DFE output rate, the number of ADC samples, the ADC Start Time, and the Tx Excess Ramp time - this varies the Ramp
    End Time.

    (3) depending on the post-processing, and the number of frames or chirps that are needed, establishes the time from data collection to
    output. This can be profiled with the CCS tools, and the mmwave SDK functions
    (a) as each Chirp data set is collected, it can be DMAed to the CSI-2/LVDS output, and sent out the LVDS output. Typically in this case,
    the Idle time is increased, and the chirp data is sent out before the next chirp starts.
    (b) if you are processing distance, each chirps' data is processed with an FFT to convert frequency to radial distance. The time it takes
    depends on the device, either the Hardware Accelerator 1443, or DSP 1642 performs the FFT - typically by the next chirp
    (c) if you are processing velocity - you typically use multiple chirps within a frame, so after the frame is collected, the Hardware
    Accelerator /MSS 1443, or DSP/MSS 1642 provides the 2nd dimension FFT Output.
    (d) if you are performing detection (software steps after the radial position and velocity calculation) these are performed on the last
    frames' data

    Q3 - see Q1

    Regards,
    Joe Quintal
  • Joe-san,

     

    Thank you for the answer.

    I understood that it is difficult to specify the time because it is depending on use-cases.

    I will answer it to the customer and if we get any additional question, please let me ask you again.

     

    Best Regards,

    Nobu Arai