• State TI Thinks Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 1 answer
  • Subscribers 31 subscribers
  • Views 283 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.

AWR1243BOOST: Channel-to-channel match and calibration

michael saxton
Prodigy 30 points
Part Number: AWR1243BOOST
Other Parts Discussed in Thread: AWR1243

While processing data collected with an AWR1243Boost we are noticing significant differences between the outputs of TI's software and our own processing.  When using TI's tools we can see outputs that match our expectations of what we should see from our data collection.  In particular, it appears all four channels are coherent and approximately gain matched.  However, when we re-process the recorded digital I/Q in our own tools the same data shows that the channel-to-channel phase match is approximately random and a 20 dB difference between channels.  It appears the device is "coherent" when run through TI's tools but incoherent when using the raw data.  It appears TI's tools are able to apply a transfer function to the raw data before processing and displaying it to correct for the errors.

A calibration data file (containing hex values) appears to have shipped with TI's software.  Does this mean that the AWR1243 does not meet the phase and gain specifications w/out the calibration data contained in that file?  We purchased the AWR1243 specifically because we are using custom DSP that would not be possible with TI's other devices or TI's software.  How do we make use of this calibration data without having to use TI's post-processing tools?  Can TI provide a method for us to interpret the data in that file and the code necessary to turn that data into the correction factors necessary to phase- and gain-match the device?

If we can't use this file, does TI have a different piece of software that would allow us to run a new calibration and derive the transfer function to correct for the phase and amplitude errors?  Can we get the "corrected" I/Q data out of TI's tools so that we can process it?

  • 0
    TI__Guru 69960 points

    Hi,

    Please provide us some time to get back to you.

    Thanks for your patience.

    Regards,

    Jitendra

  • 0
    TI__Genius 15005 points

    Hi,

    We do not include any calibration data for the AWR1243 Boost EVM. The software only contains firmware binaries for the Radar and Master subsystems of the AWR1243.

    In general, the AWR1243s are configured using mmWave Studio which provides access to all APIs via a GUI (with an optional LUA scripting interface) and in conjunction with the DCA1000 allows one to capture and post-process captured data quickly for lab evaluation. Alternatively, we provide example source code as part of mmWave DFP that allows one to programmatically configure the AWR1243. The allowed APIs in both cases are the same and both allow access to all APIs documented in the Radar Interface Control Document (ICD) that is available as part of the mmWave DFP.

    Could you please provide us more information on how you are configuring the AWR devices in the two cases? The difference might lie in the two cases.

    If possible, could you also provide a sample of the captured data in both cases for us to investigate?

    Best Regards,

    Anand

  • 0 in reply to Anand Gadiyar
    Prodigy 30 points

    One set of collected (and integrated) data is attached.  The forum's limitation on attachment sizes prevents me from uploading the raw collect.

    The four channels are not phase-matched or even close to being amplitude matched.  The image below displays the range/amplitude of the four channels individually.  None of the channels are close and there is a 20 dB difference  between two of them.  We could, of course, normalize all of the channels to the worst channel but we would then lose 20 dB of sensitivity.

  • 0 in reply to michael saxton
    TI__Genius 15005 points

    Hi Michael,

    Could you please provide us more information on how you are configuring the AWR devices in the two cases? The difference might lie in this.

    For the failing case, are you using something based off the mmwavelink example? If so, are you sure you have all APIs matched to the ones you configure in the working case? And for the working case, I assume you're using mmWave Studio?

    Also, could you capture one or two frames with a small number of chirps in both cases and attach it in compressed form here? The forum should certainly allow small files like this.

    Best Regards,

    Anand

  • 0 in reply to Anand Gadiyar
    TI__Genius 15005 points

    Hi Michael,

    Please note that the AWR1243 also has some internal calibrations that are done at RF Initialization time and the enabling or disabling of which can be configured using APIs at initialization.

    That could also explain the difference you are seeing, if you have calibrations enabled in one case and not in the other. This is why we require more information on how you are configuring the devices in the two cases.

    We have an application note (http://www.ti.com/lit/an/spracf4/spracf4.pdf

  • 0 in reply to Anand Gadiyar
    TI__Genius 15005 points

    Hi Michael,

    I wanted to check if you're still blocked on this issue?

    If so, could you please provide us more details as requested?

    Best Regards,

    Anand