Interference due to metal

Will there be substantial interferences due to metal aisle if I apply the DWM1001 Module to get the real-time location in a warehouse? If so, what are the best practices to overcome these problems?

Overcoming a metallic environment could indeed be a challenge as there would be reflections and multipaths. This is described in APS06 (attached).
The sollutions to or ways around these reflections could involve software modification , filters , number of nodes placed in the area and more.

APS006_Part2_NLOS_Operation_and_Optimizations.pdf (836.1 KB)


Thanks Leo. I’ll go through this document.

Yes. Metal, humans, concrete, masonry all stop the UWB signal on the direct path. You usually get a reception, but it bounced off something along the way. This leads to a false measurement of a path longer than the direct path. Detecting and correcting for this is complex.

The best practices are:

Have enough anchors so that there are enough anchors in direct view to solve for the position. This means you can get a good position if you use the data from just those anchors which produce good data.

Have a reliable means to identify bad data from anchors. Even if you have enough good anchors, you still have to not damage the position output with the bad data, so you have to identify the bad data. Then you need a location algorithm that can use weighting of data to upweight good anchors and dewieght the bad ones. Many geometric algorithms can’t do that, so you need something more complex than that.

The first part is a cost problem, the second one is a complex algorithm problem. We have spent man years tuning our “occlusion mitigation” system to deal with these issues, including the specific use case of warehouse tracking.

Mike Ciholas, President, Ciholas, Inc
3700 Bell Road, Newburgh, IN 47630 USA
+1 812 962 9408

1 Like