Multi-floor indoor positioning mobile application using earth's magnetic field (Case study: Universitas multimedia nusantara)

Riota, Devin Ryan and Kristanda, Marcel Bonar and Prasetiyowati, Maria Irmina (2018) Multi-floor indoor positioning mobile application using earth's magnetic field (Case study: Universitas multimedia nusantara). 2017 4th International Conference on New Media Studies (CONMEDIA).

Full text not available from this repository.

Abstract

Global Positioning System (GPS) is not the only technology used to provide location information. GPS which is able to provide accurate positioning in the outdoor environment is unable to provide the same result in an indoor environment. Earth's magnetic field based indoor positioning system utilizes the uniqueness of building's structure to determine position. An indoor positioning system is not only applied to single floor environment as the development of multistory buildings continue to increase, hence the intensity of earth's magnetic field which may deteriorate as height increases might affect the overall accuracy of an indoor positioning system. The application developed is based on Android, uses fingerprinting method along with k-nearest neighbors algorithm to compare magnetic values between database and sensor readings. A positioning filter based on previous location, direction, and magnetic values is used to filter data before applying the algorithm. Plane positioning experiment results in 1.75m of average distance error and 1.26m of standard deviation using 5 as the value of k. Floor estimation shows low accuracy when the application is used to estimate which floor the user is on, with an average accuracy of 31.25 percent on the first corridor, 34.38 percent on the second corridor and 51.56 percent on stairs with a lot less data. No evidence regarding the impact of height on accuracy and precision was found using Earth's magnetic field and k-nearest neighbor algorithm as there is no correlation between accuracy, precision, and height.

Item Type: Article
Subjects: 000 Computer Science, Information and General Works > 000 Computer Science, Knowledge and Systems > 004 Computer Science, Data Processing, Hardware
000 Computer Science, Information and General Works > 000 Computer Science, Knowledge and Systems > 005 Computer Programming
000 Computer Science, Information and General Works > 000 Computer Science, Knowledge and Systems > 006 Special Computer Methods
Divisions: Faculty of Engineering & Informatics > Informatics
Depositing User: Administrator UMN Library
Date Deposited: 07 Oct 2021 07:36
Last Modified: 07 Oct 2021 07:36
URI: https://kc.umn.ac.id/id/eprint/18571

Actions (login required)

View Item View Item