一种行人遮挡下的UWB非视距传播识别方法

doi:10.11959/j.issn.2096-3750.2023.00348

物联网学报 ›› 2023, Vol. 7 ›› Issue (4): 63-71.doi: 10.11959/j.issn.2096-3750.2023.00348

• 理论与技术 • 上一篇

一种行人遮挡下的UWB非视距传播识别方法

吴彤¹^,², 李业深¹^,², 黄镇煌¹^,², 张煜³, 张万乐¹^,², 熊轲¹^,²

¹ 北京交通大学高速铁路网络管理教育部工程研究中心，北京 100044
² 北京交通大学计算机与信息技术学院，北京 100044
³ 国网能源研究院有限公司，北京 102209

修回日期:2023-07-26 出版日期:2023-12-01 发布日期:2023-12-01
作者简介:吴彤（1997- ），女，北京交通大学计算机与信息技术学院硕士生，主要研究方向为室内定位
李业深（2000- ），男，北京交通大学计算机与信息技术学院硕士生，主要研究方向为人工智能、高铁无线通信等
黄镇煌（1998- ），男，北京交通大学计算机与信息技术学院硕士生，主要研究方向为网络智能与移动计算、联邦学习
张煜（1983- ），男，博士，国网能源研究院有限公司高级研究员，主要研究方向为边缘计算、无线协作网络和能源互联网等
张万乐（1998- ），男，北京交通大学计算机与信息技术学院博士生，主要研究方向为机器学习、移动和云计算、无线通信网络
熊轲（1981- ），男，博士，北京交通大学计算机与信息技术学院教授、副院长，主要研究方向为人工智能+5G/6G网络、无线大数据分析与处理、AI 赋能的移动网络优化设计、绿色智慧物联网、网络大数据分析、雾计算/边缘计算、室内定位、基于无线大数据的人体姿态识别等
基金资助:
中央高校基本科研业务费项目(2022JBGP003);国家自然科学基金资助项目(62071033);国家重点研发计划(2020YFB1806903)

A UWB NLOS identification method under pedestrian occlusion

Tong WU¹^,², Yeshen LI¹^,², Zhenhuang HUANG¹^,², Yu ZHANG³, Wanle ZHANG¹^,², Ke XIONG¹^,²

¹ Engineering Research Center of Network Management Technology for High-Speed Railway of Ministry of Education, Beijing Jiaotong University, Beijing 100044, China
² School of Computer and Information Technology, Beijing Jiaotong University, Beijing 100044, China
³ State Grid Energy Research Institute Co., Ltd., Beijing 102209, China

Revised:2023-07-26 Online:2023-12-01 Published:2023-12-01
Supported by:
The Fundamental Research Funds for the Central Universities(2022JBGP003);The National Natural Science Foundation of China(62071033);The National Key Research and Development Program of China(2020YFB1806903)

摘要/Abstract

摘要：

超宽带（UWB, ultrawideband）技术带宽大、抗干扰能力强、多径分辨率高，是室内定位的热点技术。然而由于室内环境复杂，UWB 信号传播不可避免会受到遮挡，产生非视距（NLOS, non-line-of-sight）传播，极大降低了UWB定位的精度。因此，准确识别出NLOS数据，将其进行剔除或矫正，对缓解定位精度下降问题有重要的作用。现有NLOS识别工作多数聚焦于墙体等建筑结构遮挡的场景，行人遮挡的场景需要进一步讨论。由于人体遮挡对信号的影响复杂且不可忽略，针对行人遮挡下的UWB非视距传播识别问题进行研究。综合比较多种机器学习方法和信号特征组合，提出了一种基于第一路径信号功率、总接收信号功率和测量距离三维特征的随机森林方法，使用较少维度的特征达到了良好的NLOS识别效果。基于不同实测数据的实验结果表明，采用所提三维特征的随机森林方法在3组不同数据集上的NLOS识别准确率分别达到了99.05%、99.32%和98.81%。

关键词: UWB, 室内定位, 非视距识别, 随机森林

Abstract:

Ultrawideband (UWB) is a hot technology for indoor positioning with large bandwidth, strong anti-interference ability, and high multipath resolution capacity.However, due to the complex indoor environment, UWB signal propagation will inevitably be blocked, resulting in non-line-of-sight (NLOS) propagation, which greatly reduces the accuracy of UWB positioning.Therefore, identifying NLOS signals accurately and discarding or correcting them are important to alleviate the problem of the decline in positioning accuracy.The majority of present NLOS identification work focuses on scenes with building structures such as walls.Further discussion is needed for scenes obscured by pedestrians.Since the impact of human obstacles on the signals is more complex and cannot be ignored, the NLOS identification under pedestrian occlusion was studied.By comparing a variety of machine learning methods and signal feature combinations, the random forest method based on the three-dimensional features of the first path signal power, the received signal power, and the measured distance was proposed.These features with fewer dimensions and easy extraction were used to achieve a high identification percentage for NLOS.The experimental results based on the measured data of different devices show that the NLOS identification accuracy based on the proposed method reaches 99.05%, 99.32% and 98.81% respectively.

Key words: UWB, indoor positioning, NLOS identification, random forest

中图分类号:

TN91

吴彤, 李业深, 黄镇煌, 张煜, 张万乐, 熊轲. 一种行人遮挡下的UWB非视距传播识别方法[J]. 物联网学报, 2023, 7(4): 63-71.

Tong WU, Yeshen LI, Zhenhuang HUANG, Yu ZHANG, Wanle ZHANG, Ke XIONG. A UWB NLOS identification method under pedestrian occlusion[J]. Chinese Journal on Internet of Things, 2023, 7(4): 63-71.

图/表 13

图1

图2

图3

图4

图5

表1

表2

表3

不同特征、机器学习方法在S1上的NLOS与LOS识别精度"

特征	naive Bayes		SVM		KNN		FNN		RF
特征	准确率	召回率	准确率	召回率	准确率	召回率	准确率	召回率	准确率	召回率
Md	53.78%	40.44%	54.82%	50.85%	91.14%	85.45%	86.00%	73.37%	90.94%	84.93%
FP、RX	76.98%	80.31%	76.73%	80.43%	79.46%	80.13%	79.53%	79.98%	79.47%	78.69%
Md、FP	76.69%	80.55%	85.20%	82.56%	98.80%	97.93%	98.33%	97.35%	98.72%	98.19%
Md、FP、RX	76.07%	81.58%	85.88%	80.37%	98.92%	98.10%	98.34%	97.40%	$\underline{99.05 %}$	$\underline{98.54 %}$

表3

表4

不同特征、机器学习方法在S2上的NLOS与LOS识别精度"

特征	naive Bayes		SVM		KNN		FNN		RF
特征	准确率	召回率	准确率	召回率	准确率	召回率	准确率	召回率	准确率	召回率
Md	56.10%	50.38%	54.77%	55.60%	90.20%	83.95%	88.33%	79.41%	90.22%	84.40%
FP、RX	70.58%	67.15%	70.90%	61.86%	77.02%	76.65%	77.04%	78.32%	77.07%	78.32%
Md、FP	70.28%	69.17%	89.28%	84.68%	99.24%	98.82%	97.85%	96.34%	99.21%	98.84%
Md、FP、RX	70.45%	69.51%	89.59%	84.57%	99.26%	98.78%	98.43%	97.38%	$\underline{99.32} %$	$\underline{98.94} %$

表4

表5

不同特征、机器学习方法在S3上的NLOS与LOS识别精度"

特征	naive Bayes		SVM		KNN		FNN		RF
特征	准确率	召回率	准确率	召回率	准确率	召回率	准确率	召回率	准确率	召回率
Md	54.52%	63.89%	53.01%	81.14%	94.15%	90.14%	93.33%	89.57%	93.92%	90.31%
FP、RX	67.90%	63.10%	82.41%	73.58%	87.26%	80.55%	86.44%	79.10%	84.70%	82.32%
Md、FP	57.87%	43.89%	72.32%	75.52%	96.32%	94.02%	88.65%	86.82%	96.23%	95.36%
Md、FP、RX	66.74%	63.35%	86.05%	81.04%	98.23%	97.27%	94.46%	91.65%	$\underline{98.81} %$	$\underline{98.54} %$

表5

表6

图6

表7

参考文献 36

[1]	BIANCHI V , CIAMPOLINI P , DE M I . RSSI-based indoor localization and identification for ZigBee wireless sensor networks in smart homes[J]. IEEE Transactions on Instrumentation and Measurement, 2019,68(2): 566-575.
[2]	MAINETTI L , PATRONO L , SERGI I . A survey on indoor positioning systems[C]// Proceedings of 2014 22nd International Conference on Software,Telecommunications and Computer Networks (SoftCOM). Piscataway:IEEE Press, 2015: 111-120.
[3]	ALARIFI A , AL-SALMAN A , ALSALEH M ,et al. Ultra wideband indoor positioning technologies:analysis and recent advances[J]. Sensors (Basel,Switzerland), 2016,16(5): 707.
[4]	FONTANA R J . Recent system applications of short-pulse ultra-wideband (UWB) technology[J]. IEEE Transactions on Microwave Theory and Techniques, 2004,52(9): 2087-2104.
[5]	ZAFARI F , GKELIAS A , LEUNG K K . A survey of indoor localization systems and technologies[J]. IEEE Communications Surveys ＆Tutorials, 2019,21(3): 2568-2599.
[6]	SOGANCI H , GEZICI S , POOR H V . Accurate positioning in ultra-wideband systems[J]. IEEE Wireless Communications, 2011,18(2): 19-27.
[7]	LIU H , DARABI H , BANERJEE P ,et al. Survey of wireless indoor positioning techniques and systems[J]. IEEE Transactions on Systems,Man,and Cybernetics,Part C (Applications and Reviews), 2007,37(6): 1067-1080.
[8]	VENKATESH S , BUEHRER R M . NLOS mitigation using linear programming in ultrawideband location-aware networks[J]. IEEE Transactions on Vehicular Technology, 2007,56(5): 3182-3198.
[9]	KHODJAEV J , PARK Y , MALIK A S . Survey of NLOS identification and error mitigation problems in UWB-based positioning algorithms for dense environments[J]. Annals of Telecommunications - Annales Des Télécommunications, 2010,65(5): 301-311.
[10]	NESSA A , ADHIKARI B , HUSSAIN F ,et al. A survey of machine learning for indoor positioning[J]. IEEE Access, 2020(8): 214945-214965.
[11]	KRISTENSEN J B , MASSANET G M , JENSEN O K ,et al. Non-line-of-sight identification for UWB indoor positioning systems using support vector machines[C]// Proceedings of 2019 IEEE MTT-S International Wireless Symposium (IWS). Piscataway:IEEE Press, 2019: 1-3.
[12]	BARRAL V , ESCUDERO C J , GARCíA-NAYA J A , ,et al. NLOS identification and mitigation using low-cost UWB devices[J]. Sensors, 2019,19(16): 3464.
[13]	MARANò S , GIFFORD W M , WYMEERSCH H ,et al. NLOS identification and mitigation for localization based on UWB experimental data[J]. IEEE Journal on Selected Areas in Communications, 2010,28(7): 1026-1035.
[14]	LI W J , ZHANG T T , ZHANG Q Y . Experimental researches on an UWB NLOS identification method based on machine learning[C]// Proceedings of 2013 15th IEEE International Conference on Communication Technology. Piscataway:IEEE Press, 2014: 473-477.
[15]	RAMADAN M , SARK V , GUTIERREZ J ,et al. NLOS identification for indoor localization using random forest algorithm[C]// Proceedings of 22nd International ITG Workshop on Smart Antennas.[S.l.:s.n.], 2018: 1-5.
[16]	JIANG C H , SHEN J C , CHEN S ,et al. UWB NLOS/LOS classification using deep learning method[J]. IEEE Communications Letters, 2020,24(10): 2226-2230.
[17]	ZHU Y P , XIA W W , YAN F ,et al. NLOS identification via AdaBoost for wireless network localization[J]. IEEE Communications Letters, 2019,23(12): 2234-2237.
[18]	OLEJNICZAK A , BLASZKIEWICZ O , CWALINA K K ,et al. Deep learning approach for LOS and NLOS identification in the indoor environment[C]// Proceedings of 2020 Baltic URSI Symposium (URSI). Piscataway:IEEE Press, 2020: 104-107.
[19]	WU C , HOU H W , WANG W J ,et al. TDOA based indoor positioning with NLOS identification by machine learning[C]// Proceedings of 2018 10th International Conference on Wireless Communications and Signal Processing (WCSP). Piscataway:IEEE Press, 2018: 1-6.
[20]	KILI? Y , ALI A J , MEIJERINK A ,et al. The effect of human-body shadowing on indoor UWB TOA-based ranging systems[C]// Proceedings of 2012 9th Workshop on Positioning,Navigation and Communication. Piscataway:IEEE Press, 2012: 126-130.
[21]	何杰, 吴雅南, 段世红 ,等. 人体对UWB测距误差影响模型[J]. 通信学报, 2017,38(S1): 58-66.
	HE J , WU Y N , DUAN S H ,et al. Model of human body influence on UWB ranging error[J]. Journal on Communications, 2017,38(S1): 58-66.
[22]	LI X F , HE J , XU L Y ,et al. The effect of multipath and NLOS on TOA ranging error and energy based on UWB[C]// Proceedings of 2016 IEEE International Conference on Consumer Electronics-Taiwan (ICCE-TW). Piscataway:IEEE Press, 2016: 1-2.
[23]	NAM S C , CHOI H B , KO Y B . On mitigation of ranging errors for through-the-body NLOS conditions using convolutional neural networks[C]// Proceedings of 2022 24th International Conference on Advanced Communication Technology (ICACT). Piscataway:IEEE Press, 2022: 141-144.
[24]	CWALINA K K , OLEJNICZAK A , BLASZKIEWICZ O ,et al. Fast fading influence on the deep learning-based LOS and NLOS identification in wireless body area networks[C]// Proceedings of 2021 34th General Assembly and Scientific Symposium of the International Union of Radio Science (URSI GASS). Piscataway:IEEE Press, 2021: 1-4.
[25]	BHARADWAJ R , ALOMAINY A , KOUL S K . Experimental investigation of body-centric indoor localization using compact wearable antennas and machine learning algorithms[J]. IEEE Transactions on Antennas and Propagation, 2022,70(2): 1344-1354.
[26]	FERREIRA A G , FERNANDES D , BRANCO S ,et al. Feature selection for real-time NLOS identification and mitigation for bodymounted UWB transceivers[J]. IEEE Transactions on Instrumentation and Measurement, 2021(70): 1-10.
[27]	CWALINA K K , RAJCHOWSKI P , BLASZKIEWICZ O ,et al. Deep learning-based LOS and NLOS identification in wireless body area networks[J]. Sensors, 2019,19(19): 4229.
[28]	WEI J Y , WANG H , SU S ,et al. NLOS identification using parallel deep learning model and time-frequency information in UWB-based positioning system[J]. Measurement, 2022(195): 111191.
[29]	KOLAKOWSKI M , MODELSKI J . First path component power based NLOS mitigation in UWB positioning system[C]// Proceedings of 2017 25th Telecommunication Forum (TELFOR). Piscataway:IEEE Press, 2018: 1-4.
[30]	GURURAJ K , RAJENDRA A K , SONG Y ,et al. Real-time identification of NLOS range measurements for enhanced UWB localization[C]// Proceedings of 2017 International Conference on Indoor Positioning and Indoor Navigation (IPIN). Piscataway:IEEE Press, 2017: 1-7.
[31]	LinkTrack datasheet[EB]. 2021.
[32]	Decawave Ltd. DWI000 USER MANUAL[EB]. 2017.
[33]	ALBAIDHANI A , MORELL A , VICARIO J L . Ranging in UWB using commercial radio modules:experimental validation and NLOS mitigation[C]// Proceedings of 2016 International Conference on Indoor Positioning and Indoor Navigation (IPIN). Piscataway:IEEE Press, 2016: 1-7.
[34]	KIM D H , KWON G R , PYUN J Y ,et al. NLOS identification in UWB channel for indoor positioning[C]// Proceedings of 2018 15th IEEE Annual Consumer Communications ＆ Networking Conference (CCNC). Piscataway:IEEE Press, 2018: 1-4.
[35]	王奕森, 夏树涛 . 集成学习之随机森林算法综述[J]. 信息通信技术, 2018,12(1): 49-55.
	WANG Y S , XIA S T . A survey of random forests algorithms[J]. Information and Communications Technologies, 2018,12(1): 49-55.
[36]	方匡南, 吴见彬, 朱建平 ,等. 随机森林方法研究综述[J]. 统计与信息论坛, 2011,26(3): 32-38.
	FANG K N , WU J B , ZHU J P ,et al. A review of technologies on random forests[J]. Statistics ＆ Information Forum, 2011,26(3): 32-38.

数据集	采集设备	距离/m	数据集大小/组
S1	LinkTrack S	1.2～7.8	72 000
S2	LinkTrack S	1.2～8.4	130 000
S3	ULM1	1.2～8.4	104 000

数据集	naive Bayes	SVM	KNN	FNN	RF
S1	0.56	48.47	0.55	17.86	5.46
S2	1.22	119.29	1.00	79.76	12.71
S3	0.29	104.86	0.52	32.24	12.81

遮挡物	准确率	召回率
门	96.04%	97.24%
墙壁	99.97%	99.94%
家具	100%	100%

一种行人遮挡下的UWB非视距传播识别方法

A UWB NLOS identification method under pedestrian occlusion

在线阅读

PDF下载

可视化

摘要/Abstract

引用本文

使用本文

图/表 13

参考文献 36

相关文章 6

Metrics

推荐阅读 0

数据集	文献[18]方法		文献[19]方法		本文方法
数据集	准确率	召回率	准确率	召回率	准确率	召回率
S1	79.53%	79.98%	54.82%	50.85%	99.05%	98.54%
S2	77.04%	78.32%	54.77%	55.60%	99.32%	98.94%
S3	86.44%	79.10%	53.01%	81.14%	98.81%	98.54%

[1]	王福展, 朱晓荣, 陈美娟, 朱洪波. 基于生成对抗网络的高精度室内无线定位方法[J]. 物联网学报, 2021, 5(2): 107-115.
[2]	单少伟,杨卫东,肖乐,王珂. Wi-Pest：一种基于CSI的储粮害虫检测方法[J]. 物联网学报, 2020, 4(4): 51-61.
[3]	常娥,冷卫杰. 工业互联网无线室内定位技术概述[J]. 物联网学报, 2020, 4(2): 129-135.
[4]	党小超,司雄,郝占军,黄亚宁,黑毅力. 复杂室内环境下基于CSI的定位方法[J]. 物联网学报, 2018, 2(4): 68-77.
[5]	庞茂,王玫,宋浠瑜,罗丽燕,阮琨. 基于声波测距与PDR融合的手机室内定位方法[J]. 物联网学报, 2018, 2(2): 58-64.
[6]	牛建伟, 齐之平, 吕卫锋, 李辉勇, 盛浩. 基于多信息融合的室内定位系统[J]. 物联网学报, 2017, 1(1): 55-66.