• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 2 answers
  • Subscribers 31 subscribers
  • Views 594 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.

AM2732-Q1: 2x AWR2243 cascading Antenna Geometry

Mohamed Amine MTAALLAH
Prodigy 200 points
Part Number: AM2732-Q1
Other Parts Discussed in Thread: TIDA-020047, AWR2243

Hello,

We used profile configuration from the application report (SWRA574B) and we found the results (point clouds) aren't coherent.

We started debugging to found the root cause and we found that the antenna geometry structure which defines our antenna is not used in any step of data processing.

How could we configure the processing chain to indicate that the second TX antenna is used for elevation  and the other for azimuth.(because with cascading we have now 6 TX (5 azimuth and 1 elevation)  and  8 RX) ?

Thanks.

  • 0
    TI__Mastermind 20605 points

    Hi Amine,

    You will have to make modifications in the following places majorly:

    1. The function DPC_ObjDet_estimateXYZ in objectdetection.c
    2. Antenna pattern related specifications of macros in objectdetection.h
    3. The function DPU_DopplerProcHWA_DDMADemod in the file dopplerprochwaDDMA.c if you are going for DDMA. The antenna samples would have to be rearranged after the demodulation according to the Tx order.
       

    Please note that these are only high-level pointers to where to make the code changes. The actual changes will depend heavily on your antenna pattern. 

    Regards,

    Kaushik

  • 0 in reply to Kaushik Gowda
    Prodigy 200 points

    Hi Gowda,

    Thanks for your reply,

    In my case i'm starting from TDM configuration. SO why I have to switch on DDM ? could you confirm that with actual sdk we can't implement the cascading with TDM configuration ?

    Regards

    Thanks

  • 0 in reply to Mohamed Amine MTAALLAH
    TI__Mastermind 20605 points

    Hi Amine,

    You do not have to switch to DDM. As I was unaware at the time, I suggested an additional change that would be required in the case of DDM which you can kindly ignore as of now. There is no limitation as to implementing TDM.

    Regards,

    Kaushik

  • 0 in reply to Kaushik Gowda
    Prodigy 200 points

    Hi and thank you again,

    please do have any recommandation/modification in case of TDM to support extraction of pts cloud base on the new  antenna geometry (TIDA020047)?

    Regards

  • 0 in reply to Mohamed Amine MTAALLAH
    TI__Mastermind 20605 points

    Hi Amine,

    Would it be possible to share your virtual antenna pattern? Any change that would be required would have to adhere to that.

    Regards,

    Kaushik

  • 0 in reply to Kaushik Gowda
    Prodigy 200 points

    Hi Gowda,

    We are using the TIDA-020047 refrence design, and we should update the code to match our antenna geomtry.

    The MMWAVE demo withi the MCU plus sdk 4.4.1.2 is using single radar and we are trying to update the demo for 2 x AWR2243 cascading radar use case.

    tell if you need further information.

    Thanks.

  • 0 in reply to Mohamed Amine MTAALLAH
    TI__Mastermind 20605 points

    Hi Amine,

    Thanks for confirming that your virtual antenna pattern also matches the TI design. I am also attaching the same for reference,

    Here, if you follow the convention TDM-MIMO chirping, your chirp sequence would be as follows:

    Chirp 1 -> Tx1 Device 1

    Chirp 2 -> Tx2 Device 1 (Elevation Antenna)

    Chirp 3 -> Tx3 Device 1

    Chirp 4 -> Tx1 Device 2

    Chirp 5 -> Tx2 Device 2

    Chirp 6 -> Tx3 Device 2

    In this particular case, there would be one overlapping antenna sample that would have to be dropped before the processing can be done. Since the antenna data is arranged appropriately based on the chirping, you should be able to go ahead into the angle estimation without additional changes. (The resulting antenna pattern is similar to that of the single chip case). If the specifics of antenna number, spacing etc, is described properly in the header files, the processing should go on without any issues is what I believe.

    So, it can be seen that the chirping sequence is also very important in this case to make the right changes. Let me know if you have anything different in your mind.

    General recommendation is that you make a MATLAB chain of the signal processing algorithm that complies with your waveform or chirping sequence design and match it with the silicon. At any stage you find a mismatch, you can look a little deeper and see what specific change might be needed.

    Regards,

    Kaushik