Tag-anchor distance peaks with OpenRTLS system

[jsgn3]

I use OpenRTLS hardware and firmware which uses the DW1000 IC.
I have one tag and three anchors, within a distance of some 6 m.
See the figure below, which shows 100 measurements without (left)
and with movement (right), for the three anchors.

When all four do not move I get three tag-anchor distances with a
std of about 2-4 cm, which is good (left). However, as soon as I move
the tag the distances are heavily disturbed. The peak-peak deviation is
up to 3 m (right).

The disturbances decrease with time, most often returning to the
desired situation of the left figure. However, the time to return might
be shorter or longer than the time the tag moves.

Any idea of what is going on here?

[color=#111111][size=small][font=Tahoma,Verdana,Arial,Sans-Serif][attachment=124][/font][/size][/color]

[dgleeson3]

Hi

This appears to be a result of the filtering they are using in this system. The filtering probably gives greater accuracy when the tag is stationary but the system has a delayed response time when the tag is moving and this results in a delay in arriving at an accurate estimate of the new location.

I hope this helps

Denis

[jsgn3]

@dgleeson3

Denis, thanks for your answer. Filtering is quite a generic term, and I
do not know how to change the filter in the OpenRTLS system. Furthermore,
I don’t understand how to explain the following with the filter solution:
if I let the tag swing like a pendulum (as a repeatable test movement),
the disturbance can disappear while the pendulum is still swinging, or it
can last until well after the swinging has ended.

[zoran_skrba]

This is not OpenRTLS forum, you may get better answers contacting them directly.