LMDD: Light-Weight Magnetic-Based Door Detection with Your Smartphone
Authors:
- Yiyang ZhaoYZYiyang Zhao
- Chen QianCQChen Qian
- Liangyi GongLGLiangyi Gong
- Zhenhua LiZLZhenhua Li
- Yunhao LiuYLYunhao Liu
- 1 September 2015
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 01903918,p. 919-928
Abstract
Doors are important landmarks for indoor positioning systems. Hence an accurate and light-weight door detection approach is highly desired. The state-of-the-art solutions are either vision based or infrastructure based, which incur nontrivial device or management cost. This paper presents a novel approach, Light-weight Magnetic-based Door Detection (LMDD), which only relies on the information from built-in sensors of a smartphone. LMDD detects a door by analyzing the change of magnetic signal and extracting special features caused by doors. It is light-weight in both computation and infrastructure cost. We have implemented a prototype of LMDD that has been installed on various Android phones. Experimental results show that LMDD achieves door detection accuracy of 74% in average, ranging from 66% to 85% in various typical environments such as offices, classrooms, residential houses, and a hospital.Keywords
This publication has 32 references indexed in Scilit:
- CrossNavi2014
- Jigsaw2014
- NomLoc: Calibration-Free Indoor Localization with Nomadic Access Points2014
- Hallway based automatic indoor floorplan construction using room fingerprints2013
- LocateMeACM Transactions on Intelligent Systems and Technology, 2013
- No need to war-drive2012
- Indoor location sensing using geo-magnetism2011
- ETOC: Obtaining robustness in component-based localization2010
- An empirical study of context in object detection2009
- Vision-Based Indoor Scene Analysis for Natural Landmark Detection2005