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.

MMWCAS-DSP-EVM: timestamp structure and sync

Part Number: MMWCAS-DSP-EVM
Other Parts Discussed in Thread: MMWCAS-RF-EVM

Hi!

We have the mmwcas-dsp-evm & mmwcas-rf-evm boards. In order to use this radar with other sensors we want to synchronize the time stamps of the radar with the others.

First, after we discover how to extract the 'timestamp' variable from the 'idx.bin' file (thanks to the information about the structure of the 'idx.bin' file in 'MatlabExamples\...\getValidNumFrames.m), we see that the 'timestamp' variable is 9 digits long, and we didn't find information about his structure.

For example in one capture the time stamp start with 73739072.0, and in other capture he start with 86400711.0

So, we want to know if there is information about the 'timestamp' structure, so we can figure out the starting time of the records. do you have any information about it?

Second, in which way we can synchronize between the radar and other sensors?

Thank you!

  • Hi,

    We are checking on this with the team internally. Kindly allow us some time to revert back on this.

    Thanks,

    Pradipta.

  • Hi Nir,

    Please confirm my understanding of your issue:

    1. You are trying to synchronize the capture of your radar with a secondary sensor.
    2. By your example, 73739072.0 & 86400711.0 are the T0 for the capture time starts of the radar and the secondary sensor respectively.

    The radar has an option for a hardware trigger which is also used to synchronize the chirping of the master and the slave devices. There is also an output pin which signals the trigger of the master device. You could potentially use this signal to obtain sync between the 2 sensors.

    One more method would be to control the trigger from the common host, if there is any, that would be controlling both your sensors. This would however have a small difference in the order of a few microseconds.

    Please let me know if this helps you or you are looking for something else.

    Regards,

    Kaushik

  • Hi!
    Thank you for the response.
    First I forgot to inform that we are using 'mmwcas-studio' for the recordings, so this is the main host for your device, there is another way to preform records?
    Second, I will check the HW trigger you mentioned, can I find it in the datasheet?
    Last, I want to exact my example about the time stamp- the numbers I mentioned was one of the time stamps in 2 different recordings from your device- in one recording from your device the time stamp was 73739072 and in another recording from your device the time stamp was 86400711, and that is why we can't figure out the difference between them and to translate the time stamp to a reality time indicator (like date-time value).
  • Hi Nir,

    1. mmWave Studio is currently the intended tool provided for this purpose. You can also use a demo data capture application provided as part of the Processor Radar SDK. (PROCESSOR-SDK-RADAR Software development kit (SDK) | TI.com). Has docs within it for how to get started.
    2. Sure. You can read up about it in the data sheet and also browse through e2e forum to get more information about the same.
    3. I do not think there is a way to get to know the exact time difference between 2 captures (in the order of microseconds that is). The best approximation you can get is the timestamp from the Studio's output window which will give you information about when the TDA capture command was sent out.

    May I know as to why it isn't possible to collect data in the same capture? This will have a common timestamp.

    Regards,

    Kaushik