Smartphone-based traveled distance estimation using individual walking patterns for indoor localization

Jiheon Kang, Joonbeom Lee, Doo-Seop Eom

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

We introduce a novel method for indoor localization with the user’s own smartphone by learning personalized walking patterns outdoors. Most smartphone and pedestrian dead reckoning (PDR)-based indoor localization studies have used an operation between step count and stride length to estimate the distance traveled via generalized formulas based on the manually designed features of the measured sensory signal. In contrast, we have applied a different approach to learn the velocity of the pedestrian by using a segmented signal frame with our proposed hybrid multiscale convolutional and recurrent neural network model, and we estimate the distance traveled by computing the velocity and the moved time. We measured the inertial sensor and global position service (GPS) position at a synchronized time while walking outdoors with a reliable GPS fix, and we assigned the velocity as a label obtained from the displacement between the current position and a prior position to the corresponding signal frame. Our proposed real-time and automatic dataset construction method dramatically reduces the cost and significantly increases the efficiency of constructing a dataset. Moreover, our proposed deep learning model can be naturally applied to all kinds of time-series sensory signal processing. The performance was evaluated on an Android application (app) that exported the trained model and parameters. Our proposed method achieved a distance error of <2.4% and >1.5% on indoor experiments.

Original languageEnglish
Article number3149
JournalSensors (Switzerland)
Volume18
Issue number9
DOIs
Publication statusPublished - 2018 Sep 18

Fingerprint

walking
Smartphones
Walking
Recurrent neural networks
Learning
Neural Networks (Computer)
Labels
Time series
Signal processing
learning
dead reckoning
Sensors
Efficiency
Costs and Cost Analysis
estimates
Smartphone
Costs
fixing
Experiments
signal processing

Keywords

  • Indoor localization
  • Smartphone-based pedestrian dead reckoning
  • Stride length estimation
  • Time-series signal deep learning framework

ASJC Scopus subject areas

  • Analytical Chemistry
  • Atomic and Molecular Physics, and Optics
  • Biochemistry
  • Instrumentation
  • Electrical and Electronic Engineering

Cite this

Smartphone-based traveled distance estimation using individual walking patterns for indoor localization. / Kang, Jiheon; Lee, Joonbeom; Eom, Doo-Seop.

In: Sensors (Switzerland), Vol. 18, No. 9, 3149, 18.09.2018.

Research output: Contribution to journalArticle

@article{93107044d4904b639329110923e9aabe,
title = "Smartphone-based traveled distance estimation using individual walking patterns for indoor localization",
abstract = "We introduce a novel method for indoor localization with the user’s own smartphone by learning personalized walking patterns outdoors. Most smartphone and pedestrian dead reckoning (PDR)-based indoor localization studies have used an operation between step count and stride length to estimate the distance traveled via generalized formulas based on the manually designed features of the measured sensory signal. In contrast, we have applied a different approach to learn the velocity of the pedestrian by using a segmented signal frame with our proposed hybrid multiscale convolutional and recurrent neural network model, and we estimate the distance traveled by computing the velocity and the moved time. We measured the inertial sensor and global position service (GPS) position at a synchronized time while walking outdoors with a reliable GPS fix, and we assigned the velocity as a label obtained from the displacement between the current position and a prior position to the corresponding signal frame. Our proposed real-time and automatic dataset construction method dramatically reduces the cost and significantly increases the efficiency of constructing a dataset. Moreover, our proposed deep learning model can be naturally applied to all kinds of time-series sensory signal processing. The performance was evaluated on an Android application (app) that exported the trained model and parameters. Our proposed method achieved a distance error of <2.4{\%} and >1.5{\%} on indoor experiments.",
keywords = "Indoor localization, Smartphone-based pedestrian dead reckoning, Stride length estimation, Time-series signal deep learning framework",
author = "Jiheon Kang and Joonbeom Lee and Doo-Seop Eom",
year = "2018",
month = "9",
day = "18",
doi = "10.3390/s18093149",
language = "English",
volume = "18",
journal = "Sensors (Switzerland)",
issn = "1424-8220",
publisher = "Multidisciplinary Digital Publishing Institute (MDPI)",
number = "9",

}

TY - JOUR

T1 - Smartphone-based traveled distance estimation using individual walking patterns for indoor localization

AU - Kang, Jiheon

AU - Lee, Joonbeom

AU - Eom, Doo-Seop

PY - 2018/9/18

Y1 - 2018/9/18

N2 - We introduce a novel method for indoor localization with the user’s own smartphone by learning personalized walking patterns outdoors. Most smartphone and pedestrian dead reckoning (PDR)-based indoor localization studies have used an operation between step count and stride length to estimate the distance traveled via generalized formulas based on the manually designed features of the measured sensory signal. In contrast, we have applied a different approach to learn the velocity of the pedestrian by using a segmented signal frame with our proposed hybrid multiscale convolutional and recurrent neural network model, and we estimate the distance traveled by computing the velocity and the moved time. We measured the inertial sensor and global position service (GPS) position at a synchronized time while walking outdoors with a reliable GPS fix, and we assigned the velocity as a label obtained from the displacement between the current position and a prior position to the corresponding signal frame. Our proposed real-time and automatic dataset construction method dramatically reduces the cost and significantly increases the efficiency of constructing a dataset. Moreover, our proposed deep learning model can be naturally applied to all kinds of time-series sensory signal processing. The performance was evaluated on an Android application (app) that exported the trained model and parameters. Our proposed method achieved a distance error of <2.4% and >1.5% on indoor experiments.

AB - We introduce a novel method for indoor localization with the user’s own smartphone by learning personalized walking patterns outdoors. Most smartphone and pedestrian dead reckoning (PDR)-based indoor localization studies have used an operation between step count and stride length to estimate the distance traveled via generalized formulas based on the manually designed features of the measured sensory signal. In contrast, we have applied a different approach to learn the velocity of the pedestrian by using a segmented signal frame with our proposed hybrid multiscale convolutional and recurrent neural network model, and we estimate the distance traveled by computing the velocity and the moved time. We measured the inertial sensor and global position service (GPS) position at a synchronized time while walking outdoors with a reliable GPS fix, and we assigned the velocity as a label obtained from the displacement between the current position and a prior position to the corresponding signal frame. Our proposed real-time and automatic dataset construction method dramatically reduces the cost and significantly increases the efficiency of constructing a dataset. Moreover, our proposed deep learning model can be naturally applied to all kinds of time-series sensory signal processing. The performance was evaluated on an Android application (app) that exported the trained model and parameters. Our proposed method achieved a distance error of <2.4% and >1.5% on indoor experiments.

KW - Indoor localization

KW - Smartphone-based pedestrian dead reckoning

KW - Stride length estimation

KW - Time-series signal deep learning framework

UR - http://www.scopus.com/inward/record.url?scp=85053682183&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85053682183&partnerID=8YFLogxK

U2 - 10.3390/s18093149

DO - 10.3390/s18093149

M3 - Article

C2 - 30231534

AN - SCOPUS:85053682183

VL - 18

JO - Sensors (Switzerland)

JF - Sensors (Switzerland)

SN - 1424-8220

IS - 9

M1 - 3149

ER -