IWR6843: Sensor performance

Divya Teja Settimali

Part Number: IWR6843

Hi,

I'm using the customized configuration file for the sensors and we have contacted earlier that the performance is different on each sensor. We have tested this theory by taking 4 sensors and it gives 23.11 meters, 19.10 meters, 16.00 meters and 08.50 meters range for same configuration file. I file is intended to give a range of more than 20 meters on ISK for area scanner application. What could be the possible reasons for the difference? Below is the cfg file:

sensorStop
flushCfg
dfeDataOutputMode 1
channelCfg 15 7 0
adcCfg 2 1
adcbufCfg -1 0 1 1 1
profileCfg 0 60.5 35 6 69 394758 0 30 1 256 6000 0 0 48
chirpCfg 0 0 0 0 0 0 0 1
chirpCfg 1 1 0 0 0 0 0 4
chirpCfg 2 2 0 0 0 0 0 2
frameCfg 0 2 32 0 100 1 0
lowPower 0 0
guiMonitor -1 1 0 0 0 0 0
cfarCfg -1 0 2 8 4 3 0 11 0
cfarCfg -1 1 0 4 2 3 1 11 0
multiObjBeamForming -1 1 0.5
clutterRemoval -1 1
calibDcRangeSig -1 0 -5 8 256
extendedMaxVelocity -1 0
bpmCfg -1 0 0 1
lvdsStreamCfg -1 0 0 0
compRangeBiasAndRxChanPhase 0.0 0.71313 -0.07993 0.80444 0.32242 0.65005 0.43036 0.63889 0.31528 0.56631 -0.49283 0.87103 -0.20547 0.80157 -0.02173 0.72357 -0.10864 0.39642 -0.74313 0.82355 -0.56723 0.82816 -0.34360 0.71524 -0.39929
measureRangeBiasAndRxChanPhase 0 1.5 0.2
CQRxSatMonitor 0 13 15 125 0
CQSigImgMonitor 0 115 6
analogMonitor 0 0
aoaFovCfg -1 -90 90 -90 90
cfarFovCfg -1 0 0 200
cfarFovCfg -1 1 -1.61 1.61
staticBoundaryBox -25 25 0 25 -1 2
boundaryBox -25 25 0 25 -1 2
gatingParam 5 2 2 2 10
stateParam 100 1 2 500 5
allocationParam 250 100 0.1 12 0.5 2
maxAcceleration 0.1 0.1 0.1
trackingCfg 1 2 800 20 20 260 120 90
heatmapGenCfg -1 1 0 190 250 60.0 3.0 10
staticDetectionCfg -1 0 -50.0 +50.0 -20.0 20.0 0.7 6.0 0.2 4 20.0
sensorStart

Please could you take a look into it. I would be happy to hear from you.

Thanks in advance.

Regards,

Divya Teja Settimali

over 2 years ago

0 Nathan Block over 2 years ago

TI__Mastermind 37805 points

Hi Divya,

The only quantity that I would expect to vary significantly from sensor to sensor is the compRangeBiasAndRxChanPhase, which has to get measured individually for each sensor for best performance. However, this parameter will most directly affect angle measurements, not necessarily detection range. How are you defining detection range? The range of 8m-20m is much larger than I would expect from sensor to sensor. Did you do repeated trials for each sensor too?

Best,

Nate

0 Nathan Block over 2 years ago in reply to Nathan Block

TI__Mastermind 37805 points

Link to doxygen on compRangeBiasAndRxChanPhase

C:/ti/mmwave_sdk_03_05_00_04/packages/ti/demo/xwr18xx/mmw/docs/doxygen/html/index.html#Calibration_section

0 Divya Teja Settimali over 2 years ago in reply to Nathan Block

Intellectual 360 points

Hi Nathan,

Earlie when I approached the Ti with the same problem I was suggested with the vary significantly from sensor to sensor is the compRangeBiasAndRxChanPhase. IWR6843ISK: unstable behaviour with the .bin file with HCC .cfg - Sensors forum - Sensors - TI E2E support forums I tried the approach and what you said is right, compRangeBiasAndRxChanPhase is not affecting the detection range. I have tested the same configuration on multiple sensors and that is how I came up with these values. It is very tiresome and also not an ideal solution to run the .bin file on every sensor and check if every sensor we have or comes in is fine or not. I'm looking for an ideal solution or where is the problem? I'm running it on ISK and ISK sensor is a long range sensor and I used the maximum values on area scanner demo and came up with the configuration so It has to give me accurate results upto 20 meters atleast. Please take a look at this and let me the possible things I could do from my side.

Regards,

Divya Teja Settimali

0 Nathan Block over 2 years ago in reply to Divya Teja Settimali

TI__Mastermind 37805 points

Hi Divya,

Could you describe your testing procedure to me? What hardware are you using? What height are you placing the sensor at, and what angle is it sitting at? Is there an enclosure or anything in between the sensor and the person (I assume you're using the sensor to detect people)?

A video would help too. I understand it's challenging to set up and take down the radar each time, but thorough testing is the only way to develop a robust product. Perhaps if you share what the specific difficulties are with setup/teardown then we can we address with what we do to make the process easier when we do similar experiments at TI.

Best,

Nate