Deep Learning-Based PDR Scheme for Predicting Movement Changes for Smartphone Users 


Vol. 46,  No. 11, pp. 1908-1919, Nov.  2021
10.7840/kics.2021.46.11.1908


PDF Full-Text
  Abstract

The pedestrian dead reckoning (PDR) using the inertial measurement units (IMUs) inside the smartphones, calculates the movement variations track the location of indoor smartphone users, and the research is actively underway to improve the localization accuracy by solving user stride estimation and sensor drift problems along the walking paths. In this paper, we propose a deep learning-based PDR method to address the issues arising from the existing PDR schemes. The proposed PDR method pre-processes the sensor values of accelerators, geomagnetic sensors, and gyroscopes that users can obtain when walking with smartphones outside the building, and constructs the location variations computed by GPS satellite signals as the output data. By using the supervised deep neural networks with the configured data, we predict changes in the movement of smartphone users, while walking outdoors as well as indoors. We built the apps on Android OS-based Samsung Galaxy S8 smartphones, utilized the deep learning framework of Google"s TensorFlow which is easy to port to smartphones, and verified the localization performance of the proposed method.

  Statistics
Cumulative Counts from November, 2022
Multiple requests among the same browser session are counted as one view. If you mouse over a chart, the values of data points will be shown.


  Cite this article

[IEEE Style]

K. Kim and Y. Shin, "Deep Learning-Based PDR Scheme for Predicting Movement Changes for Smartphone Users," The Journal of Korean Institute of Communications and Information Sciences, vol. 46, no. 11, pp. 1908-1919, 2021. DOI: 10.7840/kics.2021.46.11.1908.

[ACM Style]

Kwan-Soo Kim and Yoan Shin. 2021. Deep Learning-Based PDR Scheme for Predicting Movement Changes for Smartphone Users. The Journal of Korean Institute of Communications and Information Sciences, 46, 11, (2021), 1908-1919. DOI: 10.7840/kics.2021.46.11.1908.

[KICS Style]

Kwan-Soo Kim and Yoan Shin, "Deep Learning-Based PDR Scheme for Predicting Movement Changes for Smartphone Users," The Journal of Korean Institute of Communications and Information Sciences, vol. 46, no. 11, pp. 1908-1919, 11. 2021. (https://doi.org/10.7840/kics.2021.46.11.1908)

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)






pISSN : 1226 - 4717
eISSN : 2287 - 3880