Elisei Ember received the Best Paper Award for his publication “Improving the Correction of NLoS-Induced Ranging Errors in UWB Systems through Enhanced Labeling” (Elisei Ember, Jesus Pestana, Michael Krisper, Michael Stocker, Kay Römer, Carlo Alberto Boano , and Pablo Corbalán Pelegrín) at the 15th IFIP Wireless and Mobile Networking Conference (WMNC) 2024 .
https://opendl.ifip-tc6.org/db/conf/wmnc2024/wmnc2024/1571055880.pdf
Elisei’s article introduces a significant improvement in the applicability of Non-line-of-sight (NLoS) error mitigation for Ultra-Wide-Band (UWB) ranging systems. Current state-of-the-art approaches use a two-stage approach: first a classifier distinguishes between Line-of-Sight (LoS) and NLoS cases, and then a regressor corrects only the NLoS cases. However, this method often overcorrects a substantial number of LoS samples (approx. 25%), leading to errors up to 1 meter, which is impractical for industrial and real-world applications.
Elisei proposes the Ranging Error Labeling (RELa) approach, which defines more suitable labels based on the severity of the ranging error. This guides the decision on whether to apply ranging error correction and determines which data should be used to train the error correction model.
With RELa, the negative impact on LoS samples is reduced to negligible levels (85-99%), while maintaining effective NLoS error correction (<3%). This results in a highly applicable approach that can be easily extended in the future.
