Fall detection with multi-domain features by a portable FMCW radar

Chuanwei Ding, Yu Zou, Li Sun, Hong Hong, Xiaohua Zhu, Changzhi Li

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

9 Scopus citations

Abstract

Fall detection is important for senior care. In order to classify fall and other fall-similar daily motions, a novel dynamic range-Doppler trajectory (DRDT) method based on a frequency-modulated continuous-wave (FMCW) radar system is proposed. Multi-domain features including temporal changes of range, Doppler, radar cross-section (RCS) and dispersion are extracted from echo signals for a subspace K-Nearest Neighbor (KNN) machine learning classifier. Extensive experiments demonstrated its feasibility and an average accuracy of 95.5% was achieved in recognizing six typical fall-similar motions.

Original languageEnglish
Title of host publication2019 IEEE MTT-S International Wireless Symposium, IWS 2019 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781728107165
DOIs
StatePublished - May 2019
Event2019 IEEE MTT-S International Wireless Symposium, IWS 2019 - Guangzhou, China
Duration: May 19 2019May 22 2019

Publication series

Name2019 IEEE MTT-S International Wireless Symposium, IWS 2019 - Proceedings

Conference

Conference2019 IEEE MTT-S International Wireless Symposium, IWS 2019
Country/TerritoryChina
CityGuangzhou
Period05/19/1905/22/19

Keywords

  • DRDT
  • FMCW
  • fall detection
  • machine learning
  • multi-domain

Fingerprint

Dive into the research topics of 'Fall detection with multi-domain features by a portable FMCW radar'. Together they form a unique fingerprint.

Cite this