Ultra-wideband Non-line-of-sight Recognition Method Based on MTF-CNN

Ultra-wideband (UWB) technology has been extensively researched due to its promising potential in indoor positioning applications. However, the complexity of indoor environment inevitably leads to obstruction of UWB signals by various objects, including walls, metal doors, and wooden panels, during transmission. This results in a non-line-of-sight (NLOS) effect in UWB signals, which in turn reduces positioning accuracy. Therefore, recognizing NLOS is crucial for enhancing the positioning accuracy of UWB systems. This paper proposes a straightforward method to identify NLOS conditions caused by various obstacles obstructing UWB signals. Firstly, the channel impulse response (CIR) of UWB signals is extracted and analyzed. Further, the CIR data are transformed into a two-dimensional feature image using a method of Markov transition field (MTF), which is then recognized by a convolutional neural network (CNN) model. Finally, experiments are conducted to validate the reliability of the proposed method in both dynamic and static scenarios. The results demonstrate that the recognition accuracy of the MTF-CNN model proposed in this paper is 98.67% in the static scenario, and 92.5% in the dynamic scenario.