吉林大学学报(工学版) ›› 2025, Vol. 55 ›› Issue (7): 2444-2454.doi: 10.13229/j.cnki.jdxbgxb.20231101

• 通信与控制工程 • 上一篇    

基于改进WKNN的CSI被动室内指纹定位方法

邵小强1,2(),马博1,2,韩泽辉1,2,杨永德1,2,原泽文1,2,李鑫1,2   

  1. 1.西安科技大学 电气与控制工程学院，西安 710054
    2.西安科技大学 西安市电气设备状态检测与供电安全重点实验室，西安 710054
  • 收稿日期:2023-10-15 出版日期:2025-07-01 发布日期:2025-09-12
  • 作者简介:邵小强（1976-），男，副教授，博士. 研究方向：室内人员定位. E-mail： shaoxq@xust.edu.cn
  • 基金资助:
    国家自然科学基金项目(52174198)

CSI Passive indoor fingerprint positioning method based on improved weighted K⁃nearest neighbor algorithm

Xiao-qiang SHAO1,2(),Bo MA1,2,Ze-hui HAN1,2,Yong-de YANG1,2,Ze-wen YUAN1,2,Xin LI1,2   

  1. 1.College of Electrical and Control Engineering，Xi'an University of Science and Technology，Xi'an 710054，China
    2.Xi'an Key Laboratory of Electrical Equipment Condition Monitoring and Power Supply Security，Xi'an University of Scienec and Technology，Xi'an 710054，China
  • Received:2023-10-15 Online:2025-07-01 Published:2025-09-12

RICH HTML

 2   

PDF (PC)

38

摘要：

针对幅值和相位构造包含干扰过多导致定位精度低的问题，提出了一种基于改进加权K最近邻算法的信道状态信息被动室内定位方法。离线阶段，采用隔离森林法，改进阈值的小波域去噪和线性变换法对采集到的信道状态信息进行预处理，将处理后的幅相信息共同作为指纹数据，构造与参考点位置信息相关的稳定指纹数据库。在线阶段，提出改进的加权K近邻算法，对估计坐标进行重复匹配，该算法在一次匹配中得到位置坐标后，求该位置坐标在K个近邻点间的欧氏距离，并使用高斯变换对K个距离值进行权重计算，完成人员的定位。分别在教室和大厅进行实验模拟测试，实验结果表明：采用本文算法约81%的测试位置误差控制在1 m以内，可以有效提高定位精度。

关键词: 室内定位, 信道状态信息, 被动定位, 改进阈值的小波域去噪, 改进的加权K近邻算法, 高斯变换

Abstract:

A passive indoor positioning method based on improved weighted K-nearest neighbor algorithm is proposed to address the problem of low positioning accuracy caused by excessive interference in amplitude and phase construction. In the offline stage， the isolation forest method is adopted， and the wavelet domain denoising and linear transformation method with improved threshold are used to preprocess the collected channel state information. The processed amplitude and phase information is used together as fingerprint data to construct a stable fingerprint database related to the reference point position information. In the online stage， an improved weighted K-nearest neighbor algorithm is proposed to repeatedly match the estimated coordinates. After obtaining the position coordinates in a single match， the algorithm calculates the Euclidean distance of the position coordinates between K-nearest neighbor points， and uses Gaussian transformation to calculate the weight of the K distance values， completing personnel positioning. Experimental simulation tests were conducted in classrooms and halls， and it was found that approximately 81% of the proposed algorithm's testing position error was controlled within 1 meter， which can effectively improve positioning accuracy.

Key words: indoor positioning, channel state information, passive positioning, improved threshold denoising in wavelet domain, improved weighted K-nearest neighbor algorithm, Gaussian transformation

中图分类号: 

  • TN92

图1

隔离森林处理前后图"

图2

阈值函数随m的变化"

图3

去噪前后的幅值"

图4

相位经线性变换过程图"

图5

指纹定位架构图"

图6

501教室（实验环境示意图）"

图7

煤炭科技楼大厅（实验环境示意图）"

图8

数据包数量对定位精度的影响"

图9

501教室（无人）数据质量的CDF"

图10

煤炭科技楼大厅数据质量的CDF"

图11

两种不同算法不同K值的平均误差"

图12

两种算法误差对比图"

图13

本文算法在不同干扰下的误差对比图"

图14

6种定位算法误差累计分布函数图"

[1] 陈锐志, 郭光毅, 陈亮, 等.室内高精度定位技术研究应用现状与发展趋势[J]. 武汉大学学报: 信息科学版, 2023, 48(10): 1591-1600.
Chen Rui-zhi, Guo Guang-yi, Chen Liang, et al. R-esearch and application status and development tre-nds of indoor high precision positioning technology[J]. Journal of Wuhan University(Information ScienceEdition), 2023,48(10):1591-1600.
[2] 张楠, 史建华, 亿吉, 等. 基于加权质心定位的井下移动目标实时跟踪方法[J]. 吉林大学学报: 工学版, 2023, 53(5): 1458-1464.
Zhang Nan, Shi Jian-hua, Yi Ji,et al. Real-time trac-king method of underground moving target based on weighted centroid location[J]. Journal of Jilin University (Engineering and Technology Edition), 2023,53(5):1458-1464.
[3] Liu J, Wang L, Guo L, et al. A research on CSI ba-sed human motion detection in complex scenarios[C]∥Proceedings of the IEEE 19th International Conference on E-Health Networking, Applications and Services, Dalian, China,2017:1-6.
[4] 王玫, 宋志远. 基于TrAdaBoost算法为内核的行人航迹推算技术[J]. 吉林大学学报: 工学版, 2023, 53(8): 2364-2370.
Wang Mei, Song Zhi-yuan. Pedestrian dead reckoningtechnology based on TrAdaBoost algorithm[J]. Journalof Jilin University (Engineering and Technology Edition), 2023, 53(8): 2364-2370.
[5] 张海军, 孙学成, 赵小虎,等. 煤矿井下UWB人员定位系统研究[J]. 工矿自动化, 2022, 48(2): 29-34.
Zhang Hai-jun, Sun Xue-cheng, Zhao Xiao-hu, et al.Research on UWB personnel positioning system in coal mine[J]. Industrial and Mining Automation,2022, 48(2): 29-34.
[6] Wei F, Wu Y, Xu S W, et al. Accurate visible lightpositioning technique using extreme learning machineand meta-heuristic algorithm[J]. Optics Communications, 2023, 532:No. 129245.
[7] 郭苏, 邱继芸, 余江涛, 等. 变电站室内导航系统蓝牙定位技术与算法研究[J]. 信息技术, 2020, 44(8): 141-145.
Guo Su, Qiu Ji-yun, Yu Jiang-tao, et al. Research on Bluetooth localization technology and algorithm of substa-tion indoor navigation system[J]. Information technology, 2020, 44(8): 141-145.
[8] Wang F, Feng J, Zhao Y, et al. Joint activity recognition and indoor localization with WiFi fingerprints[J]. IEEE Access 2019,7:80058-80068.
[9] Zhuo Y, Zhu H, Xue H, et al. Perceiving accurate CSI phases with commodity Wi-Fi devices[C]∥Proceedings of the IEEE INFOCOM—IEEE Conference on Computer Communications, Atlanta, USA, 2017: 1-9.
[10] Yang W, Gong L, Man D, et al. Enhancing the perf-ormance of indoor device-free passive localization[J]. International Journal of Distributed Sensor Networ-ks, 2015, 11(11): 79-90.
[11] Wang X Y, Gao L J, Mao S W, et al. DeepFi:deep lear-ning for indoor fingerprinting using channel stateinforma-tion[C]∥IEEE Wireless Communications and Net-working Conference, New Orleans, USA, 2015: 1666-1671.
[12] Rao X P, Li Z, Yang Y B, et al. Device-free passivewireless localization system with weighted transferable discriminative dimensionality reduction method[J]. Teleommunication Systems: Modeling, Analysis, Designand Management,2020,75(1):15-29.
[13] 王博远, 贾瑞才, 贾浩男. 高精度位置服务终端技术与应用[J].无线电工程, 2016, 46(11): 5-8.
Wang Bo-yuan, Jia Rui-cai, Jia Hao-nan. High preci-sion location service terminal technology and applica-tion[J]. Radio Engineering, 2016, 46(11): 5-8.
[14] Wu C S, Zheng Y, Zhou Z M, et al.Phase U:real-time LOS Identification with Wi-Fi[C]∥Computer Communications, Hong Kong,China, 2015:2038-2046.
[15] Fang N, Fang X W, Lu K. Anomalous Behavior detec-tion based on the isolation forest model with multi-ple perspective business processes[J]. Electronics, 2022, 11(21): No. 3640.
[16] Wu Y, Gao G, Cui C. Improved wavelet denoising by non-convex sparse regularization under double wavelet domains[J]. IEEE Access 2019, 7: 30659-30671.
[17] Zhuo Y, Zhu H, Xue H. Identifying a new non-linear CSI phase measurement error with commodity WiFi devices[C]∥Proceedings of the IEEE 22nd International Conference on Parallel and Distributed Systems(ICPADS), Wuhan, China, 2016:72-79.
[18] 邓中亮, 尹露, 唐诗浩.室内定位关键技术综述[J].导航定位与授时, 2018, 5(3): 14-23.
Deng Zhong-liang, Yin Lu, Tang Shi-hao. Overviewof key techniques of indoor positioning[J]. Navigation Positioning and Timing, 2018, 5(3): 14-23.
[19] 史林军, 戴滔, 劳文洁, 等. 基于改进KNN算法的新能源发电单元运行状态识别[J]. 电力自动化设备, 2024(5): 65-72.
Shi Lin-jun, Dai Tao, Lao Wen-jie, et al. New energypower generation unit operating state recognition bas-ed on improved KNN algorithm[J]. Power automation equipment, 2024(5): 65-72.
[1] 孙大洋,王雪莹,韩双雪,钟辉,戴江南. 虚拟锚点高维坐标的非视距识别及定位优化[J]. 吉林大学学报(工学版), 2021, 51(6): 2207-2215.
[2] 刘富, 权美静, 王柯, 刘云, 康冰, 韩志武, 侯涛. 仿蝎子振源定位机理的位置指纹室内定位方法[J]. 吉林大学学报(工学版), 2019, 49(6): 2076-2082.
[3] 马蓓, 张海林, 张兆维, 钟鸣. 基于不完全信道信息的D2D功率分配算法[J]. 吉林大学学报(工学版), 2016, 46(4): 1320-1324.
[4] 王勇，葛建华，付少忠. 脉冲噪声下的高可靠性OFDM信道估计[J]. 吉林大学学报(工学版), 2007, 37(06): 1426-1430.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
版权所有 © 《吉林大学学报(工学版)》编辑部
地址：长春市人民大街5988号　电话：+86-431-85095297　传真：+86-431-85094074　E-mail: xbgxb@jlu.edu.cn
本系统由北京玛格泰克科技发展有限公司设计开发