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.

IWR6843ISK: Object detection beyond maximum range

Part Number: IWR6843ISK
Other Parts Discussed in Thread: IWR6843

Hello,

I'm using IWR6843ISK radar module and do some tests with OOB demo. The purpose of tests is to investigate maximum range dependence of slope of the chirp.

For theoretical calculations i use formula:

IWR6843 module has maximum IF bandwidth of 10 MHz, when ADC sampling rate is set to maximum value.

For example if slope is configured to 76.9 MHz/uS then:

Rmax = 19.5 m.

I expect to not see any detected object beyond range 19.5m, but when i do test with walking human radar detects human at range > 19.5m. Screenshot of this test:

How i should interpret these results when object is detected beyond maximum theoretical range?

Full configuration of radar:

profile12.cfg

Thank you,

Rytis

  • Hello

    In order to create chirp configuration for mmWave devices we have released the Sensing Estimator tool  which you have used already. These take into account he FMCW equations and other device assumptions to create a chirp that matches the sensing needs.

    We recommend creating chirp using that tool.

    In order to Guarantee consistent performance across all devices we recommend using the IF max limits provided by the Device Data sheet and using computations based on that.

    Thank you,

    Vaibhav

  • Please help us confirm the following:

    1. SW version used on the device. Which TIrex or SDK version was used.

    2. Picture of the device marking on which this was used.

    THank you,

    Vaibhav

  • Hello,

    Than You for quick response.

    In calculations i use IFmax value from IWR6843 datasheet.

    1. SDK version - 03_05_00_04. OOB demo firmware from Industrial Toolbox 4_7_0 labs.

    2. Picture of radarMMWAVE Boost picture

    Thank You,

    Rytis

  • Hello

    The additional detection range you are seeing can be because of the roll-off region of the IF Frequency filter range. 

    We recommend using the Device Datasheet values as the limit for valid detection ranges.

    THank you,

    Vaibhav

  • Hello

    I want to clarify several things.

    The additional detection range you are seeing can be because of the roll-off region of the IF Frequency filter range. 

    In my configuration slope is 76.9 MHz/uS. Detected object distance from radar is ~23.2 m(could be seen in picture). From Rmax formula we could calculate IF frequency at this range:

      

    In IWR6843 datasheet there is line:

    1. Nyquist frequency of maximum ADC sample rate(25Msps) is 12.5 MHz and according datasheet there is at least 60 dB anti-aliasing attenuation. 11.89 MHz is really close to 12.5 MHz. Is it still possible that radar pick up IF signal of 11.89 MHz?

    We recommend using the Device Datasheet values as the limit for valid detection ranges.

    2. Recommendation is to ignore everything that is beyond theoretical Rmax? 

    Thank You,

    Rytis

  • Hello Rytis,

    1. 11.89 MHz is really close to 12.5 MHz. Is it still possible that radar pick up IF signal of 11.89 MHz?

    Consistent performance across devices can be guaranteed by following the IF max limits provided by the device data sheet and basing your calculations off that value.  The provided max values should be used for calculations and theoretical computations.

            2.  Recommendation is to ignore everything that is beyond theoretical Rmax?

    As stated above, the theoretical Rmax is your limit for valid object detection ranges.  The device may pick up objects further than the theoretical max range, but the accuracy cannot be guaranteed.  For this reason, we recommend using the device datasheet values as the limit for valid detection ranges.

    Regards,

    Brennan

  • You may also find the following document helpful:

    Programming Chirp Parameters in TI Radar Devices

    Regards,

    Brennan

  • Hello,

    Thanks Brennan and Vaibhav for response.

    For now I'm happy with these answers.

    Rytis