Publication: Real-Time Threshold-Based Fall Detection System Using Wearable IoT
Date
2022
Authors
Amir N.I.M.
Dziyauddin R.A.
Mohamed N.
Ismail N.S.N.
Zulkifli N.S.A.
Din N.M.
Journal Title
Journal ISSN
Volume Title
Publisher
Institute of Electrical and Electronics Engineers Inc.
Abstract
This paper presents a Real-Time Fall Detection System (FDS) in the form of a wearable device integrating an ADXL335 accelerometer as a fall detection sensor, and classify the falling condition based on the threshold method. This system detects the wearer's movements and analyses the result in binary output conditions of 'Fall' for any fall occurrence or 'Normal' for other activities. The transmitter or FDS-Tx which is attached to the user's garment will constantly transmit data reading to the receiver or FDS-Rx via XBee module for data analysis. Raspberry Pi as the processor in FDS-Rx provides computational resources for immediate output analysis, by using threshold method, the computed results are sent to the cloud utilizing the Wi-Fi to display the user's condition on the authority's dashboard for further action. The working conditions of the systems are validated through an experiment of 10 volunteers whose perform several activities including fall events. Based on the threshold proposed, the results showed 97% sensitivity, 69% specificity and 83% accuracy from the experiment. Thus, this system fulfilled the real-Time working condition integrating (IoT) as accordingly. � 2022 IEEE.
Description
Internet of things; Wearable technology; ADXL 335; Condition; Detection system; Fall detection; Fall detection system; IoT; Real- time; Threshold methods; Time thresholds; Wearable devices; Fall detection