• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 31 subscribers
  • Views 199 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

Original question:

AWR2544LOPEVM: Discrepancy in Velocity Measurement

AWR2544LOPEVM: Frame Count Reduction Due to Chirp Parameter Adjustments

Shlomi Shvartzman
Prodigy 120 points
Part Number: AWR2544LOPEVM

Tool/software:

Hi,

Following the linked thread, I have identified the parameters causing the number of recorded frames to decrease by a factor of 2.

I found that decreasing the value of [ramp end time + idle time] halves the number of recorded frames. For example, decreasing the ramp end time by 0.5 µs reduces the bin file size from 28 KB to 14 KB.

Based on the chirp parameters I used (shown below), the chirp time is 6.4 µs, leaving a 3.6 µs spare time for the ramp end time, so a 0.5 µs decrease should be acceptable. Increasing the frame periodicity did not resolve the issue. The decreased values are still acceptable according to your ramp timing calculator. 

Can you explain this problem?

Thanks,
Shlomi

Parameters:
- f_start = 77 GHz
- slope = 110 MHz/µs
- fs = 40 MHz
- ADC samples = 256
- idle time = 6.5 µs
- ADC valid start time = 0
- Ramp end time = 10 µs
- Rx gain & gain target = 36 dB
- VCO1
- HPF corner freqs both on 350 kHz
- N chirps = 140
- frame periodicity = 2.7 ms
- N frames = 100

  • +1
    TI__Genius 12170 points

    Hi Shlomi,

    Shlomi Shvartzman said:
    I found that decreasing the value of [ramp end time + idle time] halves the number of recorded frames.

    Can you please confirm if you changed any other parameters? The size of adc_data.bin file should not change if you are not changing number of samples, number of chirps, number of chirp loops, number of frames or number of Rx enabled. Can you share the whole configuration or the log from output window of mmwave studio? Can you check if you have changed LVDS lane configuration in DataConfig tab?

    Size of ADC data = nFrames * nChirpLoops * nChirps * nRx * nADCSamples * Sampling_type * bytes_per_sample

    Regards,

    Samhitha

  • 0 in reply to Samhitha Bobba
    Prodigy 120 points

    Hi Samhitha,

    The only change I made was decreasing the ramp end time by 0.5usec. Both LVDS lanes are enabled.

    I recommend running a simple experiment with the configuration I provided. You'll immediately notice the problem.

    Thanks,
    Shlomi

  • 0 in reply to Shlomi Shvartzman
    TI__Genius 12170 points

    Hi Shlomi,

    Thank you for letting us know about this issue. We are able to replicate this issue at our end. I will check about this internally and let you know if I get any update.

    Regards,

    Samhitha

  • +1 in reply to Samhitha Bobba
    TI__Genius 12170 points

    Hi Shlomi,

    LVDS data rate is 500Mbps. With two lanes enabled, we can transfer 1000Mbps. In your case, chirp cycle time is 16.5us when Ramp end time is 10us. 2kB (1 chirp ADC data) of data should be transferred in 16.5us. This implies that the data rate is 993Mbps. If you decrease the ramp end time, you will not have enough time to transfer the data over LVDS and stream the captured data over ethernet. It's suggested to test with relaxed timings while working with mmwave studio.

    Regards,

    Samhitha

  • 0 in reply to Samhitha Bobba
    Prodigy 120 points

    Now it works! Thank you!