Indoor dynamic environment lo calization algorithm based on transfer learning

Can LIU; Junna SHANG; Ruijiang LI; Keqiang YUE

doi:10.11959/j.issn.1000−0801.2018170

您当前的位置：

首页 >

文章列表页 >

Indoor dynamic environment lo calization algorithm based on transfer learning

research and development | 更新时间：2024-06-05

- Indoor dynamic environment lo calization algorithm based on transfer learning
- Telecommunications Science Vol. 34, Issue 8, Pages: 98-108(2018)
- 作者机构：
- 作者简介：
- 基金信息：
- DOI：10.11959/j.issn.1000−0801.2018170
  CLC： TP393
- Published Online：2018-08，
  
  Published：20 August 2018
- 稿件说明：
移动端阅览
Can LIU, Junna SHANG, Ruijiang LI, et al. Indoor dynamic environment lo calization algorithm based on transfer learning[J]. Telecommunications science, 2018, 34(8): 98-108.
DOI：

Can LIU, Junna SHANG, Ruijiang LI, et al. Indoor dynamic environment lo calization algorithm based on transfer learning[J]. Telecommunications science, 2018, 34(8): 98-108. DOI： 10.11959/j.issn.1000−0801.2018170.

摘要

传统室内指纹定位系统的精度受指纹库中参考位置节点的密度和室内环境特征等多方面因素的制约。室内环境动态变化时RSS波动较大，通常不满足同分布的假设条件，故传统指纹定位方法难以满足高精度需求。针对室内环境动态变化导致传统算法无法精准定位问题，设计并实现了一种基于室内指纹库的迁移学习动态环境定位算法，该算法采用迁移学习的思想把不同分布的数据集嵌入对齐到潜在特征空间中，从而有效缓解了环境动态变化对系统造成的不利影响。本文算法实验数据均来自于真实的环境，通过仿真得到该算法的平均定位误差是1.23m。

Abstract

The accuracy of the traditional indoor fingerprint lo calization system is limited bymany factors

such as the density of the reference location node in the fingerprint database and the characteristics of the indoor environment.When the indoor environment changes dynami cally

the RSS fluctuates

and usually does notmeet the assumption of the same distribution.Therefore

it was difficult to obtain high-precision requirements for conventional fingerprint positioningmethod.Aiming at the problem that the traditional algorithm couldn’t locate accurately

an algorithm based on the indoor fingerprint database was designed and implemented.The algorithm adopted the idea ofmigration learning to embed different data sets into the latent feature space

and the adverse effects of environmental changes on the system weremitigated.The simulation results show that the average positioning error of this algorithm is 1.23m.

关键词

Keywords

references

LUO C , CHENG L , CHAN M C , et al . Pallas:self-bootstrapping fine-grained passive indoor lo calization using WiFimonitors [J ] . IEEE Transactions on Mobile Computing , 2017 ,PP( 99 ):1.

朱亚萍 , 夏玮玮 , 章跃跃 , 等 . 基于RSSI和惯性导航的融合室内定位算法 [J ] . 电信科学 , 2017 , 33 ( 10 ): 99 - 106 .

ZHU Y P , XIA W W , ZHANG Y Y , et al . A hybrid indoor lo calization algorithm based on RSSI and inertial navigation [J ] . Telecommunications Science , 2017 , 33 ( 10 ): 99 - 106 .

徐小良 , 高健 , 黄河 , 等 . 基于RSS空间线性相关的WLAN位置指纹定位算法 [J ] . 电信科学 , 2017 , 33 ( 3 ): 14 - 21 .

XU X , GAO J , HUANG H , et al . Fingerprint lo calization algorithm based on linear spatial dependence of WLAN RSS [J ] . Telecommunications Science , 2017 , 33 ( 3 ): 14 - 21 .

尚俊娜 , 程涛 , 盛林 , 等 . 广义延拓插值模型在RSSI测距方法中的应用 [J ] . 传感技术学报 , 2016 , 29 ( 11 ).

SHANG J N , CHENG T , SHENG L , et al . Application of generalized extended interpolationmethod in distancemeasurement based on RSSI [J ] . Chinese Journal of Sensors and Actuators , 2016 , 29 ( 11 ).

HASSAN-ALI M , PAHLAVAN K . A new statisti calmodel for site-specific indoor radio propagation prediction based on geometric optics and geometric probability [J ] . IEEE Transactions on Wireless Communications , 2002 , 1 ( 1 ): 112 - 124 .

KING T , KOPF S , HAENSELMANN T , et al . 11 and digital compasses [C ] // ACM Workshop on Wireless Network Testbeds,Experimental Evaluation and Characterization,September 29,2006,Los Angeles,CA,USA . New York:ACM Press , 2006 : 34 - 40 .

YOUSSEF M A , AGRAWALA A , SHANKAR A U . WLAN location determination via clustering and probability distributions [C ] // IEEE International Conference on Pervasive Computing and Communications,March 23-26,2003,Fort Worth,TX,USA . Piscataway:IEEE Press , 2003 : 143 - 150 .

张伟 , 花向红 , 邱卫宁 , 等 . Wi-Fi 指纹定位的一种新组合算法 [J ] . 测绘工程 , 2017 , 26 ( 3 ): 14 - 18 .

ZHANG W , HUA X H , QIU W N , et al . A new combinatorial optimization algorithm for WiFi positioning [J ] . Engineering of Surveying andMapping , 2017 , 26 ( 3 ): 14 - 18 .

张宏刚 , 黄华 . 基于RSSI 路径损耗因子动态修正的三边质心定位算法 [J ] . 传感技术学报 , 2016 , 29 ( 11 ).

ZHANG H G , HUANG H . Dynamic correction algorithm foRtrilateral centroid lo calization based on RSSI path loss factor [J ] . Chinese Journal of Sensors and Actuators , 2016 , 29 ( 11 ).

张梦丹 , 卢光跃 , 王宏刚 , 等 . 基于线性内插法改进的室内定位算法 [J ] . 电信科学 , 2017 , 33 ( 1 ): 9 - 15 .

ZHANG M D , LU G Y , WANG H G , et al . An improved indoor location algorithm based on linear interpolation [J ] . Telecommunications Science , 2017 , 33 ( 1 ): 9 - 15 .

SOROUR S , LOSTANLEN Y , VALAEE S , et al . Joint indoor lo calization and radiomap construction with limited deployment load [J ] . IEEE Transactions on Mobile Computing , 2015 , 14 ( 5 ): 1031 - 1043 .

CAI X , CHEN L , CHEN G . Constructing adaptive indoor radiomaps for dynamic wireless environments [R ] . 2013 .

陈斌涛 , 刘任任 , 陈益强 , 等 . 动态环境中的WiFi 指纹自适应室内定位方法 [J ] . 传感技术学报 , 2015 ( 5 ): 729 - 738 .

CHEN B T , LIU R R , CHEN Y Q , et al . WiFi fingerprint based self-adaptive indoor lo calization in the dynamic environment [J ] . Chinese Journal of Sensors and Actuators , 2015 ( 5 ): 729 - 738 .

YIN J , YANG Q , NI L M . Learning adaptive temporal radiomaps for signal-strength-based location estimation [J ] . IEEE Transactions on Mobile Computing , 2008 , 7 ( 7 ): 869 - 883 .

孔港港 , 杨力 , 孙聃石 , 等 . 一种基于位置指纹定位的K-均值聚类算法的改进 [J ] . 全球定位系统 , 2016 , 41 ( 5 ): 89 - 92 .

KONG G G , YANG L , SUN D S , et al . Research on an algorithm of fingerprint location based on K-means and WKNN [J ] . GNSS World of China , 2016 , 41 ( 5 ): 89 - 92 .

戴文渊 . 基于实例和特征的迁移学习算法研究 [D ] . 上海:上海交通大学 , 2008 .

DAI W Y . Instance-based and feature-based transfer learning [D ] . Shanghai:Shanghai Jiao Tong University , 2008 .

ALI S , NOBLES P . A novel indoor location sensingmechanism for IEEE 802.11 b/g wireless LAN [C ] // The Workshop on Lo calization,March 22,2007,Hannover,Germany . Piscataway:IEEE Press , 2007 : 9 - 15 .

SUN Z , CHEN Y , QI J , et al . Adaptive lo calization through transfer learning in indoor Wi-Fi environment [C ] // Seventh International Conference onMachine Learning and Applications.IEEE Computer Society,Dec 13-18,2008,San Diego,CA,USA . Piscataway:IEEE Press , 2008 : 331 - 336 .

ZHENG V W , XIANG E W , YANG Q , et al . Transferring lo calizationmodels oveRtime [C ] // AAAI Conference on Artificial Intelligence,AAAI 2008,July 13-17,2008,Chicago,Illinois,USA . New York:ACM Press , 2008 : 1421 - 1426 .

DAI W , CHEN Y , XUE G R , et al . Translated learning:transfer learning across different feature spaces [C ] // Conference on Neural Information Processing Systems,December 8-10,2008,Vancouver,British Columbia,Canada . New York:ACM Press , 2008 : 353 - 360 .

PAN S J , SHEN D , YANG Q , et al . Transferring lo calizationmodels across space [C ] // National Conference on Artificial Intelligence,July 13-17,2008,Chicago,Illinois . New York:ACMPress , 2008 : 1383 - 1388 .

WANG H Y , ZHENG V W , ZHAO J , et al . Indoor lo calization inmulti-floor environments with reduced effort [C ] // IEEE International Conference on Pervasive Computing and Communications,March 29-April 2,2010,Mannheim,Germany . Piscataway:IEEE Press , 2010 : 244 - 252 .

施浒立 , 颜毅华 , 徐国华 . 工程科学中的广义延拓逼近法(精) [M ] . 北京 : 科学出版社 , 2005 .

SHI H L , YAN Y H , XU G H . Generalized continuation approximationmethod in engineering science(fine) [M ] . Beijing : Science PressPress , 2005 .

张志富 , 裴军 , 胡超 , 等 . 车载动中通信标跟踪的广义延拓逼近算法 [J ] . 西安电子科技大学学报(自然科学版) , 2017 , 44 ( 4 ): 112 - 117 .

ZHANG Z F , PEI J , HU C , et al . Generalized extended approximation algorithm for vehicle satellite communication on themove [J ] . Journal of Xidian University(Natural Science) , 2017 , 44 ( 4 ): 112 - 117 .

ZHOU C F , MA L , TAN X Z . Joint semi-supervised RSS dimensionality reduction and fingerprint based algorithm for indoor lo calization [R ] . 2014 .

BELKIN M , NIYOGI P , SINDHWANI V . Manifold regularization:a geometric framework for learning from labeled and unlabeled examples [J ] . Journal ofMachine Learning Research , 2006 , 7 ( 1 ): 2399 - 2434 .

BELKIN M , NIYOGI P . Laplacian Eigenmaps for dimensionality reduction and data representation [M ] . Cambridge : MIT PressPress , 2003 .

FANG S H , LIN T . Principal component lo calization in indoor WLAN environments [J ] . IEEE Transactions on Mobile Computing , 2011 , 11 ( 1 ): 100 - 110 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Multi-adversarial domain adaptation method based on feature correction

A semi-supervised transfer learning recognition method for radar compound jamming under small samples

Substructure correlation adaptation transfer learning method based on K-means clustering

Load forecasting based on knowledge flow and transfer learning

Related Author

Yong ZHANG

Haoshuang LIU

Qi ZHANG

Wenzhe LIU

Jinqiang WANG

Minhong SUN

Xianghong TANG

Zhaoyang QIU

Related Institution

School of Information Engineering, Huzhou University

School of Computer ＆Information Technology, Liaoning Normal University

School of Communication Engineering, Hangzhou Dianzi University

China Aerospace Science and Technology Group 8511 Research Institute

Zhejiang Province Key Laboratory of Smart Management ＆Application of Modern Agricultural Resources

⁰