When using the nrf_dm example in the nrf connect sdk, the range of test results varies greatly at the same distance at different times. Why?

[H555]

When using two 52833 boards for distance measurement, when testing at 60cm, the distance value obtained sometimes wobbles between 0.6-1.2m, but sometimes stabilised around a certain value. When I use the two boards again, this value will change again, making my offset value uncertain

[Elfving]

Hello,

Yeah unfortunately the estimates can wobble around a bit. I would recommend having a look at this webinar about the NDT, that also compares the different results one can get. One thing to note there is how much the different types of measurements vary in accuracy and precision, and another thing is the standard deviation one should expect. For IFFT for instance, with a median 3 filter (a filter with which you choose the center value out of 3 measurements) you should still expect a standard deviation of 80cm. 

If you are able to get good enough quality measurements then you can also get a "high precision" measurement, which has a better standard deviation, though that is not necessarily something you can allways get.

Another thing to note is that all of this is measured in perfect conditions, reflections from walls and movement from people around can modify how well this works.

Regards,

Elfving