• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 3 answers
  • Subscribers 31 subscribers
  • Views 256 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.

Original question:

MMWAVE-STUDIO: How CFAR and is performed in AWR1843 or postProC of mmWave studio?

MMWAVE-STUDIO: How to compute the azimuth axis grid in point cloud plot

Palguna Kumar Reddy Gopireddy
Prodigy 131 points
Part Number: MMWAVE-STUDIO
Other Parts Discussed in Thread: AWR1843, AWR1642

Tool/software:

I am using DCA1000 to get raw data of AWR1843. After getting the data.I used postProc of mmWave studio to see the plots. 

I have following doubts

1) In the detection and angle estimation results plot, the azimuth axis shows distance in meters. How to prepare this azimuth axis grid? Does this azimuth axis grid is representing doppler axis or angle axis of the radarCube?

2) The angle of the detected target is not shown anywhere. What is the angle of the detected objects.

  • 0
    TI__Expert 7290 points

    Hi Palguna,

    Detection and angle estimation results plot corresponds to the location of the detected object in XY plane. 

     

    You can use the matlab tool (data cursor) present at the top, to find the coordinates of the detected object.

    Palguna Kumar Reddy Gopireddy said:
    How to prepare this azimuth axis grid? Does this azimuth axis grid is representing doppler axis or angle axis of the radarCube?

    Which plot are you referring to? As mentioned earlier, detection and angle estimation plot represents range of the object.

    Refer to (+) AWR1642BOOST: Angle estimation of a target - Sensors forum - Sensors - TI E2E support forums for the angle estimation. If you want to run the whole processing chain and get the pointcloud, I suggest you to run the demo present in the SDK - C:\ti\mmwave_sdk_03_06_02_00-LTS\packages\ti\demo\xwr18xx\mmw. You can download the SDK from AWR1843 data sheet, product information and support | TI.com.

    Regards,

    Samhitha

  • 0 in reply to Samhitha Bobba
    Prodigy 131 points

    I went to AWR1642BOOST: Angle estimation of a target - Sensors forum - Sensors - TI E2E support forums and understood that the detection and angle estimation results plot represents the coordinate geometry plot shown in "file:///C:/ti/mmwave_sdk_03_06_00_00-LTS/packages/ti/demo/xwr18xx/mmw/docs/doxygen/html/index.html"

    I got that the detection and angle estimation results plot doesn't show the Z axis of the coordinate geometry plot (which you also told that detection and angle estimation results plot corresponds to the location of the detected object in XY plane.) and tan-1(Y/X) gives the angle of the target.

    I used the data cursor tool at the top of the plots window and found that the X, Y coordinates of the two detected points are (-0.75, 0.9006) and (-0.3152, 2.125). 

    So from coordinate geometry plot, it makes

    R1*cos(phi1)*cos(theta1) = -0.75, R1*cos(phi1)*sin(theta1) = 0.9006. 

    R2*cos(phi2)*cos(theta2) = -0.3152, R2*cos(phi2)*sin(theta2) = 2.125. 

    From these values I calculated that 

    theta1 = tan-1(0.9006/-0.75) = - 50.213 degrees (or 129.786 degrees since sin +ve, cos -ve) 

    theta2 = tan-1(2.125/-0.3152) = -81.562 degrees (or 98.437 degrees since sin +ve, cos -ve) 

    R1*cos(phi1) = -0.75/cos(129.786) = 1.171998 (or) 0.9006/sin(129.786) = 1.171998

    R2*cos(phi2) = -0.3152/cos(98.437) = 2.148249 (or) 2.125/sin(98.437) = 2.148249

    From the figure in the question the range of taget1 is shown as 1.172, taget2 is shown as 2.1484, which is almost equal to the values calculated above.

    Thanks this cleared most of my doubts. By azimuth axis I mean't X axis in the figure. Some small doubts are still left.

    1) Does the theta values I calculated above, are same as the angle value obtained by Angle FFT? (I just got this doubt because I got different angle values for when I have done angle FFT on the detected points. So just wanted to check.)

    2) should I consider R1*cos(phi1) = 1.171998 and use it to find phi1 value by taking phi1 = cos-1(1.171998/1.172) 

                (or) 

                 consider R1 = 1.171998

          i.e. basically, I am just asking does the range of target shown in detection results is R (or) R*cos(phi), which can be used to find the Z coordinate, by finding phi value of the detected target. (or) We can't find phi value?

    3) I don't see the figure "Figure wx" in "file:///C:/ti/mmwave_sdk_03_06_00_00-LTS/packages/ti/demo/xwr18xx/mmw/docs/doxygen/html/index.html", only coordinate geometry plot figure I can see. Is it moved to some other place mmwave_sdk_03_06_00_00-LTS? Can you post that picture as it may give some more information regarding the topic.

    Also just wanted to say the scatter plot in mmWave SDK shows the points in X,Y,Z directions. The demo only runs shows that pointcloud. We neither can't get the point cloud data from it nor it shows the processingChain used.

  • 0 in reply to Palguna Kumar Reddy Gopireddy
    TI__Expert 7290 points

    Hi Palguna,

    I will check your reply and get back to you in 2-3 days.

    Regards,

    Samhitha

  • 0 in reply to Palguna Kumar Reddy Gopireddy
    TI__Expert 7290 points

    Hi Palguna,

    Palguna Kumar Reddy Gopireddy said:
    Does the theta values I calculated above, are same as the angle value obtained by Angle FFT?

    I suggest you to go through mmWave radar sensors  | TI.com. 

    For the second question, 

    x = rangeEst.*cos(azimEstDeg).cos(elevEstDeg)
    y = rangeEst.*sin(azimEstDeg).cos(elevEstDeg)
    z = rangeEst.*sin(elevEstDeg)
    Palguna Kumar Reddy Gopireddy said:
    I don't see the figure "Figure wx" in "file:///C:/ti/mmwave_sdk_03_06_00_00-LTS/packages/ti/demo/xwr18xx/mmw/docs/doxygen/html/index.html", only coordinate geometry plot figure I can see.

    Figure wx that is mentioned in the E2E is for AWR1642. Following is the screenshot from AWR1642 demo documentation.

    Regards,
    Samhitha
  • 0 in reply to Samhitha Bobba
    Prodigy 131 points

    Hi. Thanks for posting the "figure wx"

    1) I watched the mmWave radar sensors  | TI.com. It only tells what angle-FFT and why to perform angle-FFT. It won't tell whether theta values I calculated above, are same as the angle value obtained by Angle FFT? In fact, I performed just like told in the module 5 of mmWave radar sensors  | TI.com, but got different angles. Are they both should be same?

    2) What you told is already mentioned above. Are you telling I should consider R1*cos(phi1) = 1.171998 instead of R1 = 1.171998?

  • 0 in reply to Palguna Kumar Reddy Gopireddy
    TI__Expert 7290 points

    Hi Palguna,

    Palguna Kumar Reddy Gopireddy said:
    basically, I am just asking does the range of target shown in detection results is R (or) R*cos(phi),

    Range is R, so z = R*sin(elevEstDeg). You know the value of R and azimEstDeg, so you should be able to calculate the value of elevEstDeg and Z coordinate.

    Palguna Kumar Reddy Gopireddy said:
    I watched the mmWave radar sensors  | TI.com. It only tells what angle-FFT and why to perform angle-FFT. It won't tell whether theta values I calculated above, are same as the angle value obtained by Angle FFT? In fact, I performed just like told in the module 5 of mmWave radar sensors  | TI.com, but got different angles. Are they both should be same?

    I cannot comment why it's not same as I don't know how you are computing 3DFFT. Ideally you should get the same results.

    Regards,

    Samhitha