Design of IoT Platform for Elderly Care Utilizing LSTM Fall Detection Algorithm
Yufan Guo, Yantao He, Chunkai Wang, Junjie Hu, Haonan Xu, Jiesheng Zhu, Shuhua Li · 2024
This paper introduces an innovative IoT-based elderly care platform, utilizing a Long Short-Term Memory (LSTM) recurrent neural network fall detection algorithm. The proposed platform offers novel strategies for enhancing the quality and safety of elderly care services. Employing Internet of Things (IoT) technology, the platform gathers data from sensors integrated with smart wearable devices and home monitoring terminals in real time. This data is then consolidated through a central processor, integrating LoRa, Wifi communication technology with cloud computing and edge computing technologies to facilitate swift data processing and management. In relation to fall detection, the system employs a threshold analysis method coupled with the LSTM algorithm, thereby achieving precise identification of fall behavior among the elderly population.