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.

AWR1642BOOST: Calculations clarity is need in Demo visualizer output.

Part Number: AWR1642BOOST

Dear,

I have used the record feature on the mmWave Demo Visualizer which has created a 1 MB .DAT file and parse the captured data as mentioned in the document . I am able to get Range index , Doppler index , X, Y and Z values of detected points . From Doppler index I tried to calculate Velocity by using below formula .

velocity in m/s = doppler index * ( speed of light(3e8) / (2 * (startFreq * 1e9) * ((idleTime + rampEndTime) * 1e-6) * numChirpsPerFrame) ) ;

Where i got values

dopplerindex = 65534;

startFreq =77;

idletime =15;

rampendtime =150.86;

numChirpsPerFrame=32;

when i calculated the Velocity by substituting remaining values i  was getting velocity as 0.26 m/s . This Velocity values (0.26) is being constant for walking as well as running person which is not true .

Q1: Please tell where i am doing in calculation and Do i need to convert the Doppler index value in other format ?

Q2: What is the Doppler FFT Size ? Please tell me number of fft size . Because the Fs is 6.25 MHz and  it is also written in SDK user guide the maximum fft size is 1023 when i calculated bin resolution (Fs/1023)=6.1095e3 ; It means this 6.1095e3 doppler frequency is corresponds to 12 m/s velocity which shows that this Sensor will detects the target with 12m /s velocity  other than this it wont be able to detect like targets with 0.4 meter per sec velocity . Please tell me that my interpretation is correct or not .

  

  • Hi,

    It looks like you have done great progress! Congratulations!

    Can you please share the code you have used to parse the recorded data?

    It may benefit other members of the community and they may be double check if it is correct.

    The Doppler FFT size is the number of chirps (if they number of chirps is multiple of 2). It is computed by fft32x32(). So it is also possible to double check the code

    In the data above the doppler index = 65534 seems to me too large. Please review the definition of doppler index

    file:///C:/ti/mmwave_sdk_02_01_00_04/packages/ti/demo/xwr16xx/mmw/docs/doxygen/html/index.html

    The doc mentions that doppler index is signed. I think 65534 is a negative 16bit number

    Please review this information

    Thank you

    Cesar

  • Dear,

    Hello here is the code i have used to parse the recorded data 

    Parser_script.zip

    Q: If the Doppler fft Size is 32. for the Fs is 6.25 MHz i calculated bin resolution (Fs/32)=195.31e3 ; It means this 195.31 e3 doppler frequency is corresponds to 380 m/s velocity which shows that this Sensor will detects the target with 380 m /s velocity  other than this it wont be able to detect like targets with 0.4 meter per sec velocity . Please validate your answer with an example and reply as soon as possible .

  • Dear,

    Please reply to this question. 

  • Naveen,

    From the Doxygen that Cesar referenced, the actual Doppler index is not 65534, but rather -2. This is because the doppler index value is signed 16-bit value. This would mean that the value seen in memory is 0xFFFE. When converted, this is -2.

    Using the calculation you posted I arrive at the following velocity:

    velocity in m/s = doppler index * ( speed of light(3e8) / (2 * (startFreq * 1e9) * ((idleTime + rampEndTime) * 1e-6) * numChirpsPerFrame) ) 

    velocity in m/s = (-2*(2.998*10^8))/(2*(77*10^9)*((15+150.86)*10^-6)*32)

    velocity in m/s = -0.7335830093

    However, I would need to verify this against your configuration file (*.cfg) to ensure all the values were calculated correctly.

    The negative sign on the velocity value means that the object is moving towards the sensor. A positive value would indicate that an object is moving away from the sensor.

    The FFT size is determine by the ADC sample size and it is not 32. The fft32x32() function that is provided as part of the SDK means that the function accepts 32-bit values, not that the FFT size is 32.

    The BSS firmware on our mmWave Devices requires at least 64 ADC samples per chirp.

    For your reference, on the AWR1642BOOST, a typical maximum velocity seen in our reference Short Range Radar (SRR) Demo is 25 m/s.

    Regards,
    Kyle