• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 31 subscribers
  • Views 1141 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.

DCA1000EVM: CFAR Settings for mmwave studio

Mustafa Homsi
Expert 1171 points
Part Number: DCA1000EVM

Hello,

We are trying to replicate the settings of one of the chirps listed under the industrial toolbox database inside a mmwave studio environment to do further analysis. The chirp is listed in this website and also copied here for reference

http://dev.ti.com/tirex/#/?link=Software%2FmmWave%20Sensors%2FIndustrial%20Toolbox%2FChirp%20Database%2FPeople%20Tracking%20and%20Counting%20Application

dfeDataOutputMode 1
channelCfg 15 3 0
adcCfg 2 1
adcbufCfg 0 1 1 1 
profileCfg 0 77 30 7 62 0 0 60 1 128 2500 0 0 30
chirpCfg 0 0 0 0 0 0 0 1
chirpCfg 1 1 0 0 0 0 0 2
frameCfg 0 1 128 0 50 1 0
lowPower 0 1
guiMonitor 1 1 0 0
cfarCfg 6 4 4 0 0 16 16 4 4 50 62 0
doaCfg 600 1875 30 1
SceneryParam -6 6 0.05 4
GatingParam 4 3 2 0
StateParam 10 5 10 100 5
AllocationParam 350 0.01 20 1 2
VariationParam 0.289 0.289 1.0
trackingCfg 1 2 250 20 200 50 90
sensorStart

We were able to replicate all the settings related to the chirp/frame parameters but we are confused about the CFAR parameters and here are few questions:

1) Is there a way to do 2D CFAR filter in mmwave studio?
2) Which type of CFAR we should choose (CA, GO or SO)?
3) If we understand the above command correctly, then tenth parameter specify the range detection threshold but 50dB seems to be way too high

Thanks!
Mustafa

  • 0
    TI__Mastermind 28055 points

    Hi Mustafa,

    1.  While the TI Design mentions a 2D CFAR, it is really a 2 pass CFAR, which is 2 1D CFAR searches (Range and Azimuth).  Unfortunately, you will not be able to replicate the People Counting detection chain in mmWaveStudio because:

    • There is no Capon Beamforming AoA function implemented in mmWaveStudio. People Counting uses this, and it is necessary to replicate the detection chain in People Counting
    • The People counting demo does Range-azimuth detection - mmWaveStudio does Range-Doppler detection, so your results will be very difficult.

    2. The People Counting demo uses SO (Smallest Of) CFAR. 

    3. The CFAR module in People Counting uses linear CFAR, so the threshold values are different.  Unfortunately, there is no documentation that goes in depth on the CFAR in place; you will need to check the code to see the exact difference and calculate a similar dB threshold. You can look at the detection algorithm at C:\ti\mmwave_industrial_toolbox_3_3_0\labs\lab0015_pplcount_68xx\src\common\modules\detection\CFAR\src

    You will need to implement a signal chain that can read the raw data from mmWaveStudio if you want to make any significant changes to the processing.

    Regards,

    Justin

  • 0 in reply to Justin Curewitz
    Expert 1171 points
    Thank you Justin. It sounds that I may need to create my own signal chain in matlab to process the ADC data directly.
    Regarding the Capon Beamforming AoA approch implemented in people counting demo, I couldn't find a good documentation on this method. The SDK did a great job documenting the range-doppler detection approach but I couldn't find anything similar for the capon beamforming method. Would you mind letting me know if I a missing anything here or if you can recommend any good sources. Thanks! Mustafa
  • 0 in reply to Mustafa Homsi
    TI__Mastermind 28055 points
    Hi Mustafa,

    Currently, TI does not have a Matlab Source code or documentation online describing the Capon BF chain (besides the TID, which does not go into the mathematical technique.) You can look at the C src for a better understanding of the technique involved.

    Regards,
    Justin