Research Article
WiFi/PDR Integrated System for 3D Indoor Localization
@INPROCEEDINGS{10.1007/978-3-030-00557-3_44, author={Mu Zhou and Maxim Dolgov and Yiyao Liu and Yanmeng Wang}, title={WiFi/PDR Integrated System for 3D Indoor Localization}, proceedings={Machine Learning and Intelligent Communications. Third International Conference, MLICOM 2018, Hangzhou, China, July 6-8, 2018, Proceedings}, proceedings_a={MLICOM}, year={2018}, month={10}, keywords={Wi-Fi fingerprinting PDR Extended Kalman Filter Multi-floor positioning}, doi={10.1007/978-3-030-00557-3_44} }- Mu Zhou
Maxim Dolgov
Yiyao Liu
Yanmeng Wang
Year: 2018
WiFi/PDR Integrated System for 3D Indoor Localization
MLICOM
Springer
DOI: 10.1007/978-3-030-00557-3_44
Abstract
In recent years, location-based services LBS have received extensive attention from scholars at home and abroad, and how to obtain location information is a very important issue. The creation of systems for solving problems of positioning and navigation inside buildings is a very perspective, actual and complicated task, especially in a multi-floor environment. To improve the indoor localization performance, we proposed a three-dimensional (3D) indoor localization system integrating WiFi/Pedestrian Dead Reckoning (PDR), where extended Kalman filter (EKF) is used to estimate target location. The algorithm first relies on MEMS in our mobile phones to evaluate the speed and heading angle of the test nodes. Second, for two-dimensional (2D) localization, the speed and heading angle as with as the results of the WiFi Fingerprint-based localization are utilized as the inputs to the EKF. Third, the proposed algorithm works out the height of the test nodes by utilize a barometer and geographical data which have been recorded in real time. Our experimental results in a real multi-layer environment indicate that the proposed WiFi/PDR integrated system algorithm means that the localization accuracy error is at least 1 m lower than WiFi and PDR itself.