Fingerprinting localization of cross-temporal transferred and multi-source wireless signal fusion

doi:10.7523/j.issn.2095-6134.2021.06.012

Abstract

Abstract: To address the problem of limited coverage area of wireless signals and difficulty in dealing with time-varying characteristics of wireless signals in traditional fingerprinting localization, we propose a method of using multi-source wireless signals for fingerprinting localization, and the accuracy of positioning, which is affected by the time varying of signals, is mitigated by geodesic flow kernel. Firstly, we construct our datasets by a multi-round random sampling of multiple wireless signal sources, which provides a richer and more diverse fingerprint features. Secondly, we fuse geodesic flow kernels from fingerprint features of different times, so that we extend transferring methods from two domains to multiple domains. Finally, base classifiers are trained on multiple datasets, and the predicted position are obtained from all base classifiers, so as to elevate the generalization of the model. Simulation results show that the proposed method outperforms the traditional approaches in terms of positioning accuracy.

Key words: fingerprinting localization, multi-source fusion, transfer learning, ensemble learning, geodesic flow kernel

CLC Number:

SHI Daheng, LIU Ligang, ZHOU Bin, BU Zhiyong. Fingerprinting localization of cross-temporal transferred and multi-source wireless signal fusion[J]. Journal of University of Chinese Academy of Sciences, 2021, 38(6): 817-824.

References

[1] Bahl P, Padmanabhan V N. RADAR:an in-building RF-based user location and tracking system[C]//Proceedings IEEE INFOCOM 2000. Conference on Computer Com-munications. Nineteenth Annual Joint Conference of the IEEE Computer and Communications Societies (Cat. No.00CH37064). March 26-30, 2000, Tel Aviv, Israel. IEEE, 2000:775-784.
[2] Roos T, Myllymäki P, Tirri H, et al. A probabilistic appr-oach to WLAN user location estimation[J]. International Journal of Wireless Information Networks, 2002, 9(3):155-164.
[3] 杨如民, 陈敏, 余成波. 基于贝叶斯概率优化的Wi-Fi室内定位算法[J]. 计算机应用与软件, 2021, 38(2):97-102,144.
[4] Brunato M, Battiti R. Statistical learning theory for location fingerprinting in wireless LANs[J]. Computer Networks, 2005, 47(6):825-845.
[5] 李梦梦. 基于机器学习的WIFI室内定位技术研究[D].西安:西安科技大学, 2020.
[6] Wang X Y, Gao L J, Mao S W, et al. DeepFi:deep learning for indoor fingerprinting using channel state information[C]//2015 IEEE Wireless Communications and Networking Con-ference(WCNC). March 9-12, 2015, New Orleans, LA, USA. IEEE, 2015:1666-1671.
[7] 王玉环. 基于深度学习的RSSI室内指纹定位研究[D].北京:北京交通大学, 2020.
[8] 俞敬. 基于卷积神经网络的新型CSI室内定位方法研究[D]. 天津:天津工业大学, 2019.
[9] Wang B, Chen Q Y, Yang L T, et al. Indoor smartphone localization via fingerprint crowdsourcing:challenges and app-roaches[J]. IEEE Wireless Communications, 2016, 23(3):82-89.
[10] 季玉凤. 基于众包的室内Wi-Fi指纹地图构建方法[D].江苏镇江:江苏大学, 2020.
[11] 章裕润,吴飞,毛万葵. 基于移动众包的地磁传感器阵列室内指纹定位技术[J]. 传感器与微系统, 2019, 38(12):36-39.
[12] Kim Y, Chon Y, Cha H.Smartphone-based collaborative and autonomous radio fingerprinting[J]. IEEE Transactions on Systems, Man, and Cybernetics, Part C(Applications and Reviews), 2012, 42(1):112-122.
[13] 邓英. 基于多传感器信息融合的室内定位方法研究[D]. 重庆:重庆理工大学,2020.
[14] 汪振,陆奎.基于粒子滤波融合Wi-Fi和PDR的室内定位研究[J]. 电脑知识与技术, 2020, 16(9):39,42.
[15] 宋世铭,王继,韩李涛.一种基于改进粒子滤波算法的室内融合定位方法[J]. 导航定位学报, 2020, 8(1):99-106.
[16] 刘勋. 基于多维信息融合的位置指纹室内定位算法研究[D]. 河北秦皇岛:燕山大学, 2020.
[17] 韩冷,戴鹏,阳媛,等.多源信息融合的室内定位方法[J].传感器与微系统, 2020, 39(7):21-24.
[18] 段林甫. 多源异构融合定位方法研究[D]. 成都:电子科技大学, 2018.
[19] 赵万龙. 多源融合定位理论与方法研究[D]. 哈尔滨:哈尔滨工业大学,2018.
[20] Zheng V W, Xiang E W, Yang Q, et al. Transferring locali-zation models over time[C]//Proceedings of the Twenty-Third National Conference on Artificial Intelligence, Chicago, Illinois, AAAI, 2008, 3:1421-1426.
[21] Sorour S, Lostanlen Y, Valaee S, et al. Joint indoor locali-zation and radio map construction with limited deployment load[J]. IEEE Transactions on Mobile Comupting, 2015, 14(5):1031-1043.
[22] Pan S J, Tsang I W, Kwok J T, et al. Domain adaptation via transfer component analysis[J]. IEEE Transactions on Neural Networks, 2011, 22(2):199-210.
[23] Long M S, Wang J M, Ding G G, et al. Transfer feature learning with joint distribution adaptation[C]//2013 IEEE International Conference on Computer Vision. December 1-8, 2013, Sydney, NSW, Australia. IEEE, 2013:2200-2207.
[24] Wang J D, Chen Y Q, Hao S J, et al. Balanced distribution adaptation for transfer learning[C]//2017 IEEE International Conference on Data Mining(ICDM). November 18-21, 2017, New Orleans, LA, USA. IEEE, 2017:1129-1134.
[25] Gopalan R, Li R N, Chellappa R. Domain adaptation for object recognition:an unsupervised approach[C]//2011 Inte-rnational Conference on Computer Vision. November 6-13, 2011, Barcelona, Spain. IEEE, 2011:999-1006.
[26] Gong B Q, Shi Y, Sha F, et al. Geodesic flow kernel for unsupervised domain adaptation[C]//2012 IEEE Conference on Computer Vision and Pattern Recognition. June 16-21, 2012, Providence, RI, USA. IEEE, 2012:2066-2073.
[27] Ramdas A, Trillos N G, Cuturi M. On Wasserstein two-sample testing and related families of nonparametric tests[J]. Entropy, 2017, 19(2):47.
[28] Popleteev A. Indoor positioning using ambient radio signals:data acquisition platform for a long-term study[C]//2016 13th Workshop on Positioning, Navigation and Communications(WPNC). October 19-20, 2016, Bremen, Germany. IEEE, 2016:1-5.