Recognition of Blocking Categories for UWB Positioning in Complex Indoor Environment

Area of Science:

• Signal Processing
• Wireless Communications
• Indoor Positioning

Background:

• Accurate positioning relies on recognizing non-line-of-sight (NLOS) states to mitigate errors.
• Existing methods often overlook the impact of occlusion categories on spatial perception.
• Ultra-wideband (UWB) systems require detailed environmental information for precise indoor localization.

Purpose of the Study:

• To develop an advanced algorithm for identifying NLOS states and classifying occlusion categories.
• To enhance spatial information perception in UWB systems by incorporating blocking information.
• To improve the accuracy and robustness of positioning systems.

Main Methods:

• Proposed a bidirectional search algorithm based on maximum correlation, minimum redundancy, and minimum computational cost (BS-mRMRMC).
• Determined optimal channel impulse response (CIR) features for identifying LOS/NLOS states and blocking categories using constraint thresholds.
• Designed a hierarchical decision tree support vector machine (DT-SVM) using vector projection for classification.

Main Results:

• The BS-mRMRMC algorithm effectively identified NLOS/LOS states and various blocking categories.
• The DT-SVM classifier demonstrated high recognition accuracy for each occlusion category.
• Achieved an average recognition accuracy of 96.7% for occlusion categories, outperforming other algorithms.

Conclusions:

• The proposed BS-mRMRMC algorithm enhances indoor spatial perception by identifying occlusion categories.
• The DT-SVM verification confirms the algorithm's superior accuracy in UWB positioning.
• This approach offers a significant improvement for NLOS error mitigation and positioning accuracy.