AWR1642BOOST: Range profile conversion used in mmw demo visualizer.

Maruthi Machani

Part Number: AWR1642BOOST
Other Parts Discussed in Thread: AWR1642

Hi,

Please can you provide the conversion used for plotting the range profile from the AWR 1642BOOST.

I have written a standalone C code to read the range profile data from AWR 1642BOOST and used the below conversion to get the actual values of range profile.

log2Qformat = pow(2.0F,9.0F);
log2linScale = 1/(log2Qformat*NumRxAntenna*NumTxAntenna);

RangeValue = pow(2.0F,gSpectrum2[i]*log2linScale); //gSpectrum2[i] is range value received from AWR1642 boost serial port.

Regards,

Maruthi.

Below are my profile config settings:

sensorStop
flushCfg
dfeDataOutputMode 1
channelCfg 15 3 0
adcCfg 2 1
adcbufCfg 0 0 1 1
profileCfg 0 77 3 3 56 0 0 8 1 256 5000 0 0 30
chirpCfg 0 0 0 0 0 0 0 1
chirpCfg 1 1 0 0 0 0 0 2
frameCfg 0 1 64 0 100 1 0
lowPower 0 0
guiMonitor 1 1 1 0 0 1
cfarCfg 0 0 8 4 4 0 5000
cfarCfg 1 0 8 4 4 0 5000
peakGrouping 1 1 1 1 224
multiObjBeamForming 0 0.5
calibDcRangeSig 1 -5 8 256
sensorStart

over 8 years ago

0 Anil Mani over 8 years ago

TI__Expert 4720 points

Hi Maruthi

Use 'RangeValue = gSpectrum2[i]'. Alternatively, you can go to 'developer mode' on chrome and read the implementation of the visualizer GUI in javascript.

Regards,
Anil

0 Maruthi Machani over 8 years ago in reply to Anil Mani

Intellectual 390 points

Hi Anil,

In the "C:\ti\mmwave_sdk_01_00_00_05\packages\ti\demo\xwr16xx\mmw_InedaGui\docs\doxygen\html\index.html" file it was mentioned that xwr1642 sends range data in log2 magnitude.

In XWR16xx the points represent the sum of log2 magnitudes of received antennas, expressed in Q8 format.

Please can you help in decoding, I don't know java.

Regards,

Maruthi.

0 Anil Mani over 7 years ago in reply to Maruthi Machani

TI__Expert 4720 points

Hi Maruthi,

Apologies for the late reply. The formula that you gave seems to be correct.

log2Qformat = pow(2.0F,9.0F);
log2linScale = 1/(log2Qformat*NumRxAntenna*NumTxAntenna);

RangeValue = pow(2.0F,gSpectrum2[i]*log2linScale); //gSpectrum2[i] is range value received from AWR1642 boost serial port.

Regards
Anil

0 Maruthi Machani over 7 years ago in reply to Anil Mani

Intellectual 390 points

Thanks.

0 Alessandro over 7 years ago in reply to Anil Mani

Prodigy 180 points

Hi Anil,

I think that instead of
log2Qformat = pow(2.0F,9.0F)
it should be:
log2Qformat = pow(2.0F,8.0F).
The rangeProfile data are in the Q8 format, not in the Q9 format.

Best regards

Alessandro

0 Maruthi Machani over 7 years ago in reply to Alessandro

Intellectual 390 points

Hi Anil,

Is QFormat fixed or varying depending on the input data, profile configurations and from demo to demo.

Regards,

Maruthi.

0 Anil Mani over 7 years ago in reply to Maruthi Machani

TI__Expert 4720 points

Hi Maruthi, Alessandro

The QFormat (in the SDK) is a variable, and is based on the 'range resolution'. The QFormat increases as the 'range resolution' improves, in order to preserve more accurate range information.

The formula that is used in the SDK is given below.
dataPathObj->xyzOutputQFormat = (uint8_t) ceil(log10(16. /
fabs(dataPathObj->rangeResolution))/log10(2));

Regards
Anil

0 Maruthi Machani over 7 years ago in reply to Anil Mani

Intellectual 390 points

Thanks Anil.

0 Maruthi Machani over 7 years ago in reply to Maruthi Machani

Intellectual 390 points

Hi Anil,

Whatever you are telling is for range(x,y,z) QFormat.
Do we need to take this QFormat for log-magnitude spectrum also.

Regards,
Maruthi.

0 Alessandro over 7 years ago in reply to Anil Mani

Prodigy 180 points

Hi Anil,

the doxygen-documentation states the following:

"@subsubsection tlv2 Range profile
* Type: (@ref MMWDEMO_OUTPUT_MSG_RANGE_PROFILE)
*
* Length: (Range FFT size) x (size of uint16_t)
*
* Value: Array of profile points at 0th Doppler (stationary objects).
* In XWR16xx the points represent the sum
* of log2 magnitudes of received antennas, expressed in Q8 format.
* In XWR14xx the points represent the average of log2 magnitudes of
* received antennas, expressed in Q9 format."

Best regards

Alessandro

0 Alessandro over 7 years ago in reply to Alessandro

Prodigy 180 points

It's definitely in Q8-Format as you can see also from parseCfg.js in the DemoVisualizer-Tool.

if (platform == mmwInput.Platform.xWR14xx) {
    P.log2linScale[idx] = 1/512;
    if (P.dataPath[idx].numTxElevAnt == 1) P.log2linScale[idx] = P.log2linScale[idx]*4/3; //MMWSDK-439
} else if (platform == mmwInput.Platform.xWR16xx) {
    P.log2linScale[idx] = 1/(256*P.dataPath[idx].numRxAnt*P.dataPath[idx].numTxAnt);
}

0 Maruthi Machani over 7 years ago in reply to Alessandro

Intellectual 390 points

Thanks Alessandro,
To convert to linear scale we need to use pow(2.0F,gSpectrum2[i]*log2linScale);

0 Maruthi Machani over 7 years ago in reply to Alessandro

Intellectual 390 points

Thanks Alessandro, QFormat 8 is working properly.

0 Alessandro over 7 years ago in reply to Maruthi Machani

Prodigy 180 points

You are welcome. :-)

Sensors

Sensors forum

AWR1642BOOST: Range profile conversion used in mmw demo visualizer.