计算机科学

计算机科学 ›› 2024, Vol. 51 ›› Issue (5): 277-283.doi: 10.11896/jsjkx.230300137

• 计算机网络 • 上一篇    下一篇

基于迁移学习的动态环境室内定位方法研究

王佳昊1, 付一夫1, 冯海男1, 任昱衡2   

  1. 1 电子科技大学信息与软件工程学院 成都 610051
    2 白俄罗斯国立大学国际商学院 明斯克 220071
  • 收稿日期:2023-03-16 修回日期:2023-06-30 出版日期:2024-05-15 发布日期:2024-05-08
  • 作者简介:(wangjh@uestc.edu.cn)
  • 基金资助:
    电子科技大学-智小金-智能家居联合研究中心项目(H04W210180);内江市科技孵化和成果转化专项资金(2021KJFH004);四川省科学技术厅重点研发计划高新技术领域重点研发项目(2022YFG0212);四川省科技支撑项目(2021YFG0024)

Indoor Location Algorithm in Dynamic Environment Based on Transfer Learning

WANG Jiahao1, FU Yifu1, FENG Hainan1, REN Yuheng2   

  1. 1 School of Information and Software Engineering,University of Electronic Science and Technology,Chengdu 610051,China
    2 School of Business,Belarusian State University,Minsk 220071,Republic of Belarus
  • Received:2023-03-16 Revised:2023-06-30 Online:2024-05-15 Published:2024-05-08
  • About author:WANG Jiahao,born in 1978,Ph.D,associate professor,is a member of CCF(No.27769M).His main research interests include IoT,information security and data mining.
    FU Yifu,born in 1998,master,is a member of CCF(No.67870G).His mainresearch interests include indoor location,transfer learning and data mining.
  • Supported by:
    UESTC-ZHIXIAOJING Joint Research Center of Smart Home(H04W210180),Neijiang Technology Incubation and Transformation Funds(2021KJFH004), Key R & D Program Key R & D Projects in High-tech Fields(2022YFG0212) and Science and Technology Support Plan of Sichuan Province of China(2021YFG0024).

PDF (PC)

358

摘要： 随着智能家居应用的不断深化,基于Wi-Fi信号的室内定位技术也受到了广泛关注。在实际应用中,大多数室内定位算法采集得到的训练数据和测试数据通常并非来自于同一理想环境,各种环境条件变化以及信号漂移导致采集得到的训练数据和测试数据间的概率分布不同。传统定位模型在面对不同分布的训练数据和测试数据时无法保证具有良好的定位精度,常出现算法定位精度大幅降低,甚至算法不可用等问题。面对这一难点,迁移学习中的域适应方法作为一种可以有效解决训练样本和测试样本概率分布不一致的学习问题被广泛应用于室内定位领域。文中结合域适应学习和机器学习算法,提出了一种基于特征迁移的室内定位算法(Transfer Learning Location Algorithm Based on Global and Local Metrics Adaptation,TL-GLMA)。TL-GLMA在定位阶段通过特征迁移方式将两域原始数据映射至高维空间,从而在最小化两域数据的分布差异的同时保留两域数据内部的局部几何属性,并利用映射后的独立同分布数据训练分类器,从而实现目标定位。实验结果表明,TL-GLMA能够有效减少环境变化带来的干扰,提升定位精度。

关键词: 室内定位, Wi-Fi信号, 环境适应, 迁移学习, 域适应

Abstract: With the development of smart home,the Wi-Fi signal-based localization technology has also been widely studied.In actual application,the training data and test data collected by indoor positioning algorithm usually do not come from the same ideal conditions.Changes in various environmental conditions and signal drift can cause different probability distributions between the training data and test data.The existing positioning algorithm cannot guarantee stable accuracy when facing these different probability distributions,resulting in dramatic reduction and infeasibility of the positioning accuracy of indoor location algorithms.Considering these difficulties,the domain adaptation technology in transfer learning is proven to be a promising solution in past researches to solve the inconsistent probability distributions problem.In this paper,a feature transferbased indoor localization algorithm TL-GLMA is proposed by combining domain adaptation learning and machine learning algorithms.TL-GLMA maps the original data of two domains to the high-dimension space through feature transfer,so as to minimize the distribution difference between the two domains in retaining the local geometric properties.In addition,because the mapped data is independent and identically distributed,TL-GLMA can use it for training the classifier to achieve better location result.Experiment results show that TL-GLMA can effectively reduce the interference caused by environmental changes and improve the location accuracy.

Key words: Indoor location, Wi-Fi signal, Environmental adaptation, Transfer learning, Domain adaptation

中图分类号: 

  • TP393
[1]IRSAN F A S,AMINAH N S,DJAMAL M.RSSI-WIFI BasedIndoor Position Tracking System Using Support Vector Machine(SVM)[C]//2022 International Conference on Electrical,Computer,Communications and Mechatronics Engineering(ICECCME).IEEE,2022:1-5.
[2]XUE W,HUA X,LI Q,et al.A new weighted algorithm based on the uneven spatial resolution of RSSI for indoor localization[J].IEEE Access,2018,6:26588-26595.
[3]JEDARI E,WU Z,RASHIDZADEH R,et al.Wi-Fi based indoor location positioning employing random forest classifier[C]//2015 International Conference on Indoor Positioning and Indoor Navigation(IPIN).IEEE,2015:1-5.
[4]YANG Z,ZHOU Z,LIU Y.From RSSI to CSI:Indoor localization via channel response[J].ACM Computing Surveys(CSUR),2013,46(2):1-32.
[5]WANG X,GAO L,MAO S,et al.CSI-based fingerprinting for indoor localization:A deep learning approach[J].IEEE Transactions on Vehicular Technology,2016,66(1):763-776.
[6]SONG Q,GUO S,LIU X,et al.CSI amplitude fingerprinting-based NB-IoT indoor localization[J].IEEE Internet of Things Journal,2017,5(3):1494-1504.
[7]ZHANG Y,QU C,WANG Y.An indoor positioning methodbased on CSI by using features optimization mechanism with LSTM[J].IEEE Sensors Journal,2020,20(9):4868-4878.
[8]ZHANG W,LIU K,ZHANG W,et al.Deep neural networks for wireless localization in indoor and outdoor environments[J].Neurocomputing,2016,194:279-287.
[9]CHEN Z,ZOU H,YANG J F,et al.WiFi fingerprinting indoor localization using local feature-based deep LSTM[J].IEEE Systems Journal,2019,14(2):3001-3010.
[10]DANG X,TANG X,HAO Z,et al.A device-free indoor localization method using CSI with Wi-Fi signals[J].Sensors,2019,19(14):3233.
[11]CAI C,DENG L,ZHENG M,et al.PILC:Passive indoor localization based on convolutional neural networks[C]//2018 Ubi-quitous Positioning,Indoor Navigation and Location-Based Ser-vices(UPINLBS).IEEE,2018:1-6.
[12]PANS,TSANGI,KWOKJ,et al.Domain adaptation via transfer component analysis[J].IEEE Transactions on Neural Networks,2011,22(2):199-210.
[13]YIN Y,YANG X,LI P,et al.Localization with Transfer Lear-ning Based on Fine-Grained Subcarrier Information for Dynamic Indoor Environments[J].Sensors,2021,21(3):1015.
[14]LONG M,WANG J,DING G,et al.Transfer feature learning with joint distribution adaptation[C]//Proceedings of the IEEE International Conference on Computer Vision.2013:2200-2207.
[15]WANG J,CHEN Y,HAO S,et al.Balanced distribution adaptation for transfer learning[C]//2017 IEEE International Confe-rence on Data Mining(ICDM).IEEE,2017:1129-1134.
[16]SUN B,FENG J,SAENKO K.Return of frustratingly easy domain adaptation[C]//Proceedings of the AAAI Conference on Artificial Intelligence.2016.
[17]WANG J,CHEN Y,YU H,et al.Easy transfer learning by exploiting intra-domain structures[C]//2019 IEEE international conference on multimedia and expo(ICME).IEEE,2019:1210-1215.
[18]GUO X,WANG L,LI L,et al.Transferred knowledge aided positioning via global and local structural consistency constraints[J].IEEE Access,2019,7:32102-32117.
[19]LI H,ZHAO M.Indoor positioning based on hybrid domaintransfer learning[J].IEEE Access,2020,8:130527-130539.
[20]YU C,WANG J,CHEN Y,et al.Transfer learning with dyna-mic adversarial adaptation network[C]//2019 IEEE International Conference on Data Mining(ICDM).IEEE,2019:778-786.
[21]ZHU Y,ZHUANG F,WANG J,et al.Deep subdomain adaptation network for image classification[J].IEEE Transactions on Neural Networks and Learning Systems,2020,32(4):1713-1722.
[22]BORGWARDT K M,GRETTON A,RASCH MJ,et al.Integrating structured biological data by kernel maximum mean discrepancy[J].Bioinformatics,2006,22(14):e49-e57.
[23]BEN-DAVID S,BLITZER J,CRAMMER K,et al.Analysis of representations for domain adaptation[J].Advances in Neural Information Processing Systems,2006,19(1):137-144.
[24]JIANG M,HUANG W,HUANG Z,et al.Integration of global and local metrics for domain adaptation learning via dimensiona-lity reduction[J].IEEE Transactions on Cybernetics,2015,47(1):38-51.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发