Scale fingerprint clustering-based gait tracking algorithm under high-frequency gait mode switching environment

Y. Liu, S. L. Wang, L. Wang, L. L. Li

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

The MEMS inertial accelerometer has the ability to sense the acceleration information of the human body. By making use of this ability, plus the simple mode recognition algorithm, human gait could be identified from several dynamic locomotion gait modes; however, this algorithm is only effective in a low-frequency gait mode switching situation. Once the gait mode switching frequency comes to 0.05Hz or higher, its tracking accuracy drops down to 80% or less. This paper presents a gait tracking algorithm based on clustered Reference Point (RP) fingerprint, to improve the tracking accuracy under high-frequency gait mode switching situation. By running on an Android mobile phone, the new algorithm is proved by at least 13%.

Original languageEnglish
Title of host publicationEnergy, Environment and Green Building Materials - Proceedings of the International Conference on Energy, Environment and Green Building Materials, EEGBM 2014
EditorsAi. Sheng
PublisherCRC Press/Balkema
Pages281-284
Number of pages4
ISBN (Print)9781138027183
DOIs
Publication statusPublished - 2015
Externally publishedYes
EventInternational Conference on Energy, Environment and Green Building Materials, EEGBM 2014 - Guilin-Guangxi, China
Duration: 28 Nov 201430 Nov 2014

Publication series

NameEnergy, Environment and Green Building Materials - Proceedings of the International Conference on Energy, Environment and Green Building Materials, EEGBM 2014

Conference

ConferenceInternational Conference on Energy, Environment and Green Building Materials, EEGBM 2014
Country/TerritoryChina
CityGuilin-Guangxi
Period28/11/1430/11/14

Keywords

  • Clustered RP Fingerprint
  • High-frequency Gait Mode Switching
  • MEMS Inertial Accelerometer

Fingerprint

Dive into the research topics of 'Scale fingerprint clustering-based gait tracking algorithm under high-frequency gait mode switching environment'. Together they form a unique fingerprint.

Cite this