• State Resolved
  • Locked Locked
  • Replies 1 reply
  • Subscribers 32 subscribers
  • Views 880 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:

AWR1443BOOST: Validation of object's velocity

AWR1443BOOST: Which formula is correct for Doppler resolution calculation?

Chris Meng
Mastermind 35185 points
Part Number: AWR1443BOOST

Hi,

I found two formulas to calculate Doppler resolution.

One is in mmWave.js: P.dataPath[idx].dopplerResolutionMps = 3e8 / (2*P.profileCfg[profileCfgIdx].startFreq*1e9 *
                                            (P.profileCfg[profileCfgIdx].idleTime + P.profileCfg[profileCfgIdx].rampEndTime) *
                                            1e-6 * P.dataPath[idx].numChirpsPerFrame);

One is in below e2e post:

I think chirp rate = 1/(idle time+rampEndTime)

As Number of DopplerBins = Number of ChirpsPerFrame / Number of TX antennas, if number of TX antennas is not 1, the chirpsPerframe is not equal to the number of dopplerbins. That is, when no. of TX is not 1, these two formula will output different result.

Which one is correct? Or does I miss anything?

  • 0
    TI__Expert 5235 points

    The formula you quoted which has chirp rate in it is a general formula agnostic of whether the scheme is TDM-MIMO or not. For doppler estimation purposes, the chirp rate in TDM-MIMO case is not the physical chirp rate but is determined by the period between chirps corresponding to the same transmit antenna (you cannot use consecutive physical chirps that are corresponding to different tx antennas to estimate velocity because you will get phase shift due to both velocity and position). So chirp rate is (1/((idle time + rampEndTime)*numTxAntennas)). The multiplication in the denominator by number of doppler bins effectively gives you the total physical chirping time, which is what determines the doppler resolution. Although strictly speaking numDopplerBins should not be used because one could have more doppler bins (by zero padding) than what I call the number of virtual chirps (= number of physical chirps/numTxantennas) to increase accuracy but that will not increase resolution (ability to see two objects moving with velocities separated by doppler resolution). So the formula in the mmwave.js is the correct formula although the other formula in current oob implementation (where we don't allow number of virtual chirps to be non power of 2 because we don't do zero padding, this is unlike the case of range where we allow number of adc samples to be non-power of 2 and zero pad) is the same. See also