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.

AWR1443BOOST: Struggling with the complexity of SDK support, hardware choices, connectivity, and mission

Part Number: AWR1443BOOST
Other Parts Discussed in Thread: IWR1443, AWR1443

We have just received our AWR1443BOOST and mmWave DevPack boards, have reviewed the SDK docs, have mechanically assembled the two boards, and are now astonishing confused. There is not a clear, linear process to power up, USB connect, software configure, or "smoke test" the boards. Our primary goal is to establish a narrow beam distance measurement test. There seem to be many choices, but not a coherent recommendation for an initial test. Clearly, this is ignorance on our part, and if we can get pointed to a step by step process it would help. Do we plug in both micro USB connectors and the 5 volt 2.5 amp supply? What software stack gets us to operations? Are all third party run-time packages identified somewhere? In the forum I see many people asking about specific software component operations, so an over all check list would help everybody. Maybe we just didn't get the memo...

We think this is a fabulous product and we have several well defined applications for it and we look forward to success, please help!

Many thanks,

David  (Weak Minded) Drake

  • Hi David,

    Sorry to hear that you have unpleasant experience. AWR1443BOOST comes with page which has a QR code. On scanning that QR code, it takes you to the Out of box demo video. Below is that video link, you don't need to connect DevPack board for the out of box test.

    Regards,

    Bhavin

  • Hi David,
    Please let us know if you come across any more issue.
    Regards,
    Bhavin
  • Thank you gentlemen. We will give this a try!
    DD
  • Progress so far; Got the web app, got the TI Cloud Agent extension. Powered up the board and connected the micro USB to the PC. No obvious smoke. The LED lights.

    No extra ports showed up, but four "other devices", listed as

    XDS110 (02.03.00.07) Embed with CMSIS-DAP 

    did show up. Obviously there are drivers missing. Looked though the SDK docs and found no clear references.

    Can you point me to the needed drivers?

    TNX

    DD

  • Hello David,
    You can refer to the user guides to help you through the initial bringup process.
    For the hardware connections and driver installation details you can refer to the AWR1443BOOST EVM user guide ( www.ti.com/.../swru507.pdf). The section 2.5 explains the PC connection and the COM ports expected. There is also a link to the drivers to be installed if your PC does not recognize the device.

    The other user guide that would guide you through the demo application and usage is the mmwave SDK user guide: software-dl.ti.com/.../mmwave_sdk_user_guide.pdf

    Do let us know incase you still have difficulty running the demo application.

    Regards,
    Vivek
  • Hello,


    Seems like the PC is not recognizing the device ports and needs drivers installed:

    Please check the SDK user's guide  section "3. 1. Connecting the xWR14xx/xWR16xx EVM to PC"

    And the Release notes for installation of emulation package.

    Above documents can be found in SDK User's guide is available in the "docs" directory under the installation folder.

    Thank you,

    Vaibhav

  • We are making progress. We did locate the driver reference in the emulation software. The demo visualizer starts. We see COM 14 and COM 15 in the device manager and use these respectively..There was also an old COM 1. The data channel insisted on using this COM 1 rather than COM 15. We disabled the COM 1 and the visualizer retained COM 14 and COM 15 properly. The visualizer communicated with the 1443.

    We used the default config and the system started. We see a processor utilization display of about 15%. We did see a sensorStart and Done in the log.

    In the PLOT page we see no radar data displayed on the graphs. We tried a variety of potential rages and targets, but see no data. The yellow and red LEDs are "On" on the radar board.

    From the docs we think we should be seeing radar returns. Advice?

    Here is part of the console log:


    mmwDemo:/>adcbufCfg 0 0 1 1
    Done

    mmwDemo:/>profileCfg 0 77 429 7 57.14 0 0 70 1 256 5209 0 0 30
    Done

    mmwDemo:/>chirpCfg 0 0 0 0 0 0 0 1
    Done

    mmwDemo:/>chirpCfg 1 1 0 0 0 0 0 2
    Done

    mmwDemo:/>frameCfg 0 1 16 0 100 1 0
    Done

    mmwDemo:/>guiMonitor 1 1 0 0 0 1
    Done

    mmwDemo:/>cfarCfg 0 0 8 4 4 0 5120
    Done

    mmwDemo:/>cfarCfg 1 0 4 2 3 0 5120
    Done

    mmwDemo:/>peakGrouping 1 1 1 1 255
    Done

    mmwDemo:/>multiObjBeamForming 1 0.5
    Done

    mmwDemo:/>sensorStart
    Done

    DD

  • Hello David,

    Based on t your description seems like the device is communicating with the Visualizer gui:

    Commands accepted by gui -   implies Device is getting commands from Visualizer gui.

    Gui plot shows device utilization -  implies Visualizer Gui can get data from device.

    A few questions:

    Is the Cpu utilization updating as time progresses this will ensure that data is live and gui is not stuck.

    In the Plot area of visualizer is the Range plot showing FFT plot?  Please check "3.3 Range Profile"  in mmWave Demo Visualizer User Guide

    Thank you,

    Vaibhav

  • Hi Vaibhav,

    Yes, the CPU utilization is updating. All other plots are blank, regardless of targets or range. The board seems to be functioning. I have not changed any firmware on the board. All is as shipped

    Here is a screen shot:

  • More progress! I just used the heat maps only with a frame rate of one Hz and saw realistic data. Still no radar data.
    DD
  • More problems.... The USB ports 14 and 15 exist and the demo visualizer finds them, but struggles with communications. Several reboots didn't help. Any thoughts on USB stability improvement?
  • The next problem appears to be USB connectivity stability. The Visualizer will detect the Com ports 14 and 15, but sits at "waiting for data". Again many reboots and restart without success. This is on a well updated Windows 7 system. Advice?
    DD
  • It seems like the receivers are working as we get the heat map data, but the transmitters are sending no signals to receive. Is this a 1443BOOST problem, a software problem, or a dumb user problem?

    DD
  • Hello David,

    Sorry to hear that you are still seeing issues with Out of Box demo.


    In order to make sure there is nothing wrong with com port connection would you be able to try it on a different PC?
    If that doesn't work then I would recommend reflashing the bin file which is available via SDK directory and steps mentioned in user's guide.

    We will check back here in sometime to check your update today!.

    Thank you,
    Vaibhav
  • The USB problem has settled down. It just takes a certain pattern of resetting and the sacrifice of a small chicken.  We set the scatter and range plots, and the range profile came up with no data. Behold, when we reset the axies on the range  profile we started to see range data. The other plots are still blank.

    Here is the plot:    AT least the transmitters are running and we are getting some data.   I have to say that I have worked with TI products extensively since 1970 and this one has provided the most opportunity for unintentional discovery.  Any suggestions for the other plots?   The axis reset doesn't do it.

    David Drake

  • Hi David,

    Good to see the range plot going.

    I see no objects marked though and it could be the CFAR setting.

    Can you please try the "Load config from PC and send" button to send the cfg file provided as part of SDK.

    Typical path for the file would be as below:

    C:\ti\mmwave_sdk_01_00_00_05\packages\ti\demo\xwr14xx\mmw\profiles\profile_2d.cfg

    Thank you,

    Vaibhav

  • Dear Vaibhav,

    Success! The CFAR settings are actually marked as "Range and Doppler Detection Threshold" in the Visualizer UI, by setting these to 5 rather than 15, all of a sudden, like ghosts from ancient radar antiquity, dots started to appear. Their placement was logical. Very good advice.

    So we will proceed...

    David Drake
  • Hello David,

    Glad to know that CFAR setting were enough to get this to work.  CFAR thresholds set the level above which a certain frequency peak gets denoted as an object these have to be adapted to the scene. The Out of Box values work for a wide variety of scenes seems like it needed some tweaking for your scenario.

    Please feel free to check out the "CFAR Cfg" field in the mmWave SDK user's guide.

    IWR1443/AWR1443 has a CFAR hardware and select few settings of this hardware are controlled by the "CFAR Cfg" field of profile sent to visualizer or by altering the "Range and Doppler Detection Threshold" in the Visualizer interface.

    Details of IWR1443/AWR1443 CFAR hardware are covered in Radar Hardware Accelerator User's Guide - Part 2 .

    Thank you,

    Vaibhav

  •  An image where CFAR values set to 5