TITLE:
UWB NLOS Signal Recognition Based on Deep Learning
AUTHORS:
Guanhui Li, Guoliang Wei, Zhuang Xi
KEYWORDS:
Ultra-Wide Band, Non-Line-of-Sight, Deep Learning, Convolutional Neural Network, Long Short-Term Memory, Self-Attention Mechanism
JOURNAL NAME:
Open Journal of Applied Sciences,
Vol.16 No.3,
March
12,
2026
ABSTRACT: Ultra-wide band (UWB) positioning technology has attracted increasing attention due to its high ranging accuracy. However, in indoor environments, non-line-of-sight (NLOS) signals significantly degrade the ranging performance of UWB positioning systems compared with line-of-sight (LOS) signals. Therefore, accurate identification of LOS and NLOS signals is essential before adopting effective measures to improve UWB positioning accuracy. In this paper, a deep learning-based classification algorithm for UWB NLOS signal identification is proposed. Convolutional neural networks are employed to enhance the model’s feature extraction capability, while long short-term memory networks are used to capture temporal dependencies in the time-series data. Furthermore, a self-attention mechanism is introduced to further strengthen feature representation, thereby improving the recognition accuracy of NLOS signals. Experimental results demonstrate that the proposed method achieves an overall accuracy of 89.95%, with a precision and specificity of 91.07% and 91.33%, respectively, indicating superior comprehensive performance.