摘要
在超宽带(UWB)室内定位系统中,采用时域特征识别非视距信号(NLOS)不能达到令人满意的性能。为了解决这一问题,利用小波变换方便同时提取时频域特征的特点,提出了一种基于连续小波变换和卷积神经网络(CWT-CNN)的NLOS识别方法。仿真实验使用一个网络开源数据集。仿真结果表明,对于6种不同的室内场景,NLOS信号识别准确率分别为87.50%、84.50%、88.00%、87.00%、90.50%和88.50%,CWT-CNN能够较好的识别时频域内的NLOS信号。
In the ultra wide band(UWB) indoor positioning system, using time-domain features to identify non-line-ofsight(NLOS) signals cannot achieve satisfactory performance. In order to solve this problem, taking advantage of the fact that wavelet transform facilitates simultaneous extraction of features in the time-frequency domain, this paper proposes a NLOS recognition method based on continuous wavelet transform and convolutional neural network(CWTCNN). The simulation experiment uses an open network source data set. The simulation results show that in six different indoor scenes, the accuracy of NLOS signal recognition is 87.50%, 84.50%, 88.00%, 87.00%, 91.00% and 90.00%respectively and that CWT-CNN can better identify NLOS signal in the time-frequency domain.
作者
张忠健
席志红
ZHANG Zhongjian;XI Zhihong(College of Information and Communication Engineering,Harbin Engineering University,Harbin 150001,China)
出处
《应用科技》
CAS
2022年第2期81-86,共6页
Applied Science and Technology
关键词
室内定位
超宽带
非视距信号识别
连续小波变换
深度学习
卷积神经网络
支持向量机
特征可视化
indoor positioning
ultra wide band
non-line-of-sight signal recognition
continuous wavelet transform
deep learning
convolutional neural network(CNN)
support vector machine(SVM)
feature visualization
