• State Resolved
  • Locked Locked
  • Replies 14 replies
  • Answers 1 answer
  • Subscribers 32 subscribers
  • Views 1263 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.

IWR6843ISK: Increasing the range of the IWR6843ISK

Christian Paulik
Prodigy 10 points
Part Number: IWR6843ISK
Other Parts Discussed in Thread: IWR6843, , AWR1843, UNIFLASH

Hello,

We are trying to use the IWR6843ISK EVM for a medium range radar application. Specifically we want to reliably detect vehicles at 150m with velocities up to 150kmph. I found the medium range radar lab for the AWR1843 in the radar toolbox and noticed a few things. First is that it doesn't appear that the IWR6843 has enough L3 memory for the radar data cube (768kB vs the required 1024 for the given chirp configurations). I have discovered TI's memory compression and decompression engine described here: https://www.ti.com/lit/an/swra663/swra663.pdf?ts=1687955652855&ref_url=https%253A%252F%252Fwww.google.com%252F, however I have no idea how to use it. Our current system reads data over the serial COM ports and does any filtering, clustering, detection, and plotting using Python. Is there an easy, user friendly way to implement the memory compression on the board so that we can still read the serial data and use our Python scripts?

Secondly, we have tried to implement beamforming by using all three transmit antenna in unison but it wasn't helping detect out further. I found that the AWR1843 required mounting a 0-ohm resistor to supply more voltage so that the 3 Tx antenna could chirp in unison. I then looked at the long range people detection lab: file:///C:/ti/mmwave_industrial_toolbox_4_9_0/labs/long_range_people_detection/68xx_long_range_people_det/docs/68xx_long_range_people_det_user_guide.pdf

and found that the IWR also requires more power but the manual says it requires the ICBOOST. Is there a simpler way to provide sufficient power to the board in order to enable beamforming?

Thank you

  • 0
    TI__Expert 8802 points

    Hi Christian,

    Please allow us another 24 hours to provide a response.

    Regards,

    Tim

  • 0 in reply to Tim Henderson
    TI__Guru*** 148804 points

    Hi,

    With the IWR6843ISK EVM antenna you will not be able to detect reliably a car at 150m. The ISK antenna will probably be able to achieve a range of about 120m.

    Thank you

    Cesar

  • 0 in reply to Cesar
    Prodigy 10 points

    Cesar,

    While I appreciate your response, this doesn't exactly answer my two questions regarding memory compression and providing enough power for beamforming. Could you please elaborate on these points?

    Thank you,

    Christian

  • 0 in reply to Christian Paulik
    TI__Expert 4448 points

    Christian,

    You listed an older toolbox version. I would recommend downloading the newer Radar Toolbox. Then, please refer to the following guide detailing the memory compression model used on the device: <RADAR_TOOLBOX>\tools\memory_compression\radar_memory_Compression_Model_iwr6843_users_guide.pdf

    Also, for the details regarding capon beamforming technique we apply, reference <RADAR_TOOLBOX>\source\ti\examples\People_Tracking\docs

    Regards,
    Luke

  • 0 in reply to Luke Rouleau
    Prodigy 10 points

    Luke,

    From what I can tell this guide describes using a MATLAB script to compare performance between memory compression and no memory compression, but it doesn't actually tell me how to implement the memory compression on the IWR6843ISK. Am I missing something?

    Thank you,

    Christian

  • 0 in reply to Christian Paulik
    Prodigy 10 points

    Luke,

    Please disregard my previous comment, I found a configuration file command "compressCfg" which seems to do what I have been looking for.

    Best,

    Christian

  • 0 in reply to Christian Paulik
    Prodigy 10 points

    Hello,

    I have a follow up question. I would like to be able to use all 3 Tx antenna simultaneously and I know that I need to provide the IWR EVM sufficient power. Another forum post

    https://e2e.ti.com/support/sensors-group/sensors/f/sensors-forum/974809/iwr6843isk-tx-beamsteering-with-dca1000evm

    said that powering the board via the test points TP12 and TP13 would be sufficient, but another forum 

    https://e2e.ti.com/support/sensors-group/sensors/f/sensors-forum/854718/iwr1843-beam-forming-in-elevation-plus-azimuth-with-3-antennas

    mentioned an LDO bypass and PA LDO bypass that must be used. I am not sure what this second forum was talking about and I am confused. Would I be able to simply power the board from the test points or are there extra steps I must employ to give the proper power?

    Thank you,

    Christian

  • 0 in reply to Christian Paulik
    Prodigy 10 points

    Hello,

    I have done some more research on the LDO bypass and PA  LDO disable settings. It appears these can be configured using mmwave studio, but before I do so I want to make sure I don't damage my EVM. There are some warnings concerning setting these configurations and that they may damage the EVM, however it appears that in order to properly use all 3 Tx antenna simultaneously the bypass must be enabled. If I power my EVM board via the test points TP12 and TP13 with a 5V 3A supply and enable the bypass, will this be sufficient to provide enough power for all 3 Tx, or will it damage my board?

    Thank you,

    Christian

  • 0 in reply to Christian Paulik
    TI__Expert 4448 points

    Hi Christian,

    Please allow me some time to get the right eyes on this question for you. Meanwhile, can you please elaborate on why you want to transmit off of all 3 Tx at the same time? 

    Regards,
    Luke

  • 0 in reply to Luke Rouleau
    Prodigy 10 points

    Hi Luke,

    Thank you for your timely response. In the long range people detection lab, beamforming is used with all 3 Tx simultaneously to increase the range of the radar. I would like to do something similar for vehicle detection. While Cesar stated above that 150m would not be possible, I would still like to push my EVM to the limit and see how far I can detect vehicles reliably. 

    Best,

    Christian

  • 0 in reply to Christian Paulik
    TI__Mastermind 25835 points

    Christian,

        Current EVMs on hardware configuration comes with internal LDO bypass option taken into consideration. That is, PMIC/DCDC provides 1V input at the LDO inputs, Hence in the mmWave studio or SDK internal LDOs are bypassed so that 1V would be directly fed to RF circuits. Concern is only during 1.3V being fed to LDO input and internal LDOs are bypassed, then 1.3V will be directly would be fed to RF circuits.

    Hence you could use all 3 Tx simultaneously to increase the range of the Radar. There is no concern of EVM getting damaged with the same default configuration. 

    Thanks and regards,

    CHETHAN KUMAR Y.B. 

  • 0 in reply to CHETHAN KUMAR Y.B.
    Prodigy 10 points

    Chethan,

    Thank you for your response that was very helpful! However, now I have ran into another problem. In the mss_main.c file located in the directory,

    C:\ti\mmwave_sdk_03_06_00_00-LTS\packages\ti\demo\xwr68xx\mmw\mss

    I found a line of code to enable the LDO bypass 

    I set the .ldoBypassEnable line equal to 3 which should enable the LDO bypass and disable the PA LDO which seems to be what I want according to the mmWave Radar Interface Control Document 

    I then rebuilt the binaries and flashed the board via UNIFLASH, but now I am experiencing some strange behavior. Connecting the board to TI's Demo Visualizer works fine, I am able to send my custom config and observe data on the plots, but when I try to send my custom config via my python script, it doesn't seem to be working and I receive no radar data (this script was running perfectly fine before I made this change in the mss_main.c and reflashed the board). I also noticed that the D7 LED is illuminated extremely dimly when pressing the reset button and does not illuminate at all whilst running my python script, however when running on the Demo Visualizer the LED shines brightly as it should. Do you have any insight as to what is happening? Is there something I might have missed?

    Thank you,

    Christian

  • 0 in reply to Christian Paulik
    Prodigy 10 points

    Chethan,

    You can disregard my last comment (although if you confirmed I did things correctly that would be helpful) but I appear to have it working now!

    Thank you,

    Christian

  • +1 in reply to Christian Paulik
    TI__Mastermind 25835 points

    Christian,

         It appears your configurations are correct for the LDO bypass.

    Thanks and regards,

    CHETHAN KUMAR Y.B.