DW1001C Module obstacle Test Accuracy

Wireless Connectivity Ultra-Wideband
, ,
1

Hello,

I am using DWM1001 Module in that LOS i am getting 10 Cm Accuracy which is very impressive ,

But if some obstacle is coming in between Initiator and Responder then Accuracy is Reducing to 50 cm - 80 cm. Device is Showing 50 cm - 80 cm More then Actual Distance.

Can you suggest some way , How i can Get Possible Accurate Location ?

Thanks & Regards,
Hardik

2

Easy. Move the obstacle out of the way.

In all seriousness it does depend on the type of obstacle. If itโ€™s something that will reflect/block the signal, e.g. something metallic or with a high water content like a person, then about all you can do is move things so the line of sight isnโ€™t blocked.
The system measures range by measuring the time of flight of a radio signal and assuming that it a) has traveled in a straight line and b) travelled at the speed of light. If the direct signal canโ€™t get through then it will end up measuring the distance along the shortest reflected path off a nearby wall, floor, ceiling or anything else near.

If the obstacle isnโ€™t completely blocking the signal but is just reducing its strength to the point that the DW1000 isnโ€™t detecting the direct signal then it may be possible to get the correct distance by playing with some of the internal registers to adjust the detection thresholds. You should be able to tell whether this is the case or not by looking at the CIR data. Generally anything along these lines will require you to run your own firmware on the modules to some extent. If this is possible then the measurement accuracy will still be decreased slightly, weaker signals mean more noise and the signal will travel at a different speed through the object breaking the other assumption made in the calculations.

Other possible approaches are to measure to look at signal levels and attempt to detect and discard indirect signals or to measure to more points than are required and then detect and discard measurements that donโ€™t fit.

1 Like
3

Thanks Andy For Replyโ€ฆ

I am developing Social Distance Application where i can not Move Obstacle.

Anyway to Measure Signal Strength at Receiver Side ? Any Register or Something?

Thanks & Regards,
Hardik

4

Hi Hardik,

I suggest looking at APS006, more specifically Part 2: Non Line of sight operation.

The โ€œOther possible approaches are to measure to look at signal levels and attempt to detect and discard indirect signals or to measure to more points than are required and then detect and discard measurements that donโ€™t fit.โ€ AndyA mentioned are explained in more detail here. Basically you want to adjust the receiver detection threshold and reject false first path detections. As AndyA mentioned, this requires you to create your own firmware. PANS is designed as a fairly simple RTLS demo and does not have NLOS support (or supports it very limited).

5

Thanks Seppe,

I am trying to create my Own Firmware only , Thanks For clarification , I will try from my end.

Thanks ,
Hardik

6

Hello Seppe,

Using 0x12 RX Frame Quality Information .

As Explain in DW1000 User Manual . โ€œRX_POWER โ€“ FP_POWER, is less than 6dB the channel is likely to be LOS, whilst if the difference is greater than 10dB the channel is likely to be NLOS.โ€.

And this scenario is working fine , But Sometime even though there is obstacle in between two devices Signal is Reaching with Correct Signal Parameter + CIR .

Using this i am Estimating RX_LEVEL.

๐‘…๐‘‹ ๐ฟ๐‘’๐‘ฃ๐‘’๐‘™ = 10 ร— log10 (๐ถ ร— 217/ ๐‘2) โˆ’ ๐ด ๐‘‘๐ต๐‘š

And also I am monitoring CIR power Separately .

So any other Parameter or Way to Confirm , whether Any Obstacle is there or not ?

Or How to Avoid this Special Case , Where , With Obstacle , TOF is increasing , RX_LEVEL and CIR is Reaching with Perfect Value. ?

Thanks & Regards,
Hardik