Open AccessResearch on the Recognition of Abnormal Behaviors in the Elderly Based on Wi-Fi Signals
Author(s) -
Yifan Li,
Yongchun Cao,
Qiang Lin,
WeiQiong Wang
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1848/1/012077
Subject(s) - elderly people , disease , set (abstract data type) , abnormality , computer science , pattern recognition (psychology) , artificial intelligence , medicine , gerontology , pathology , psychiatry , programming language
Wandering behaviour is an important diagnostic indicator for patients with Alzheimer’s disease, and falls are the main threat to the health of the elderly. Effective detection of these two types of abnormal behaviours of the elderly and timely intervention are important for improving the quality of life of patients with Alzheimer’s disease and related diseases significance. This study proposes and studies a method for detecting and identifying abnormal behaviours of the elderly based on Wi-Fi signals. First, the correlation between Wi-Fi signal changes and abnormal behaviours of the elderly was analyzed; secondly, this study applied general Wi-Fi equipment to obtain the status data of the abnormal behaviours of the elderly and preprocess the data through Hampel filtering; then, the effective features of the CSI sequence fragments were designed and extracted, a data set of 2243 samples including 7 types of abnormal behaviours of the elderly was constructed; finally, a BiLSTM-Based abnormal behaviour recognition model for the elderly was constructed, and the average classification accuracy reached 95.95%, which proved this experiment is feasible and effective.