TY  - JOUR
T1  - Smartphone User Location Initialization in Multi-Floor Indoor Localization Using Magnentic Field Map and Deep Learning Model
AU  - Kim, Jin-Woo 
AU  - Cho, Sanghoon 
AU  - Shin, Yoan 
JO  - The Journal of Korean Institute of Communications and Information Sciences
PY  - 2024
DA  - 2024/1/1
DO  - 10.7840/kics.2024.49.12.1698
KW  - Smartphone-based indoor localization
KW  - Deep Learning
KW  - Magnetic field map
KW  - 3D location initialization
KW  - pedestrian dead reckoning
AB  - In this paper, we propose a 3D location initialization method that combines a magnetic field map and a deep learning model to detect the initial floor and 2D location of a smartphone user, addressing the challenges of multi-floor indoor localization. The proposed method meets the requirements of infrastructure-free multi-floor indoor localization, and experiments have validated its performance in estimating the user's 3D position.