Key points are not available for this paper at this time.
Impairment and a substantial decline in the mobility, independence, and quality of life of an elderly person. In this regard, the current work suggests a novel IoT-based system that makes the use of low-power wireless sensing the networks, big data, cloud computing and smart devices to detect falls of older persons in interior situations. A Three-dimensional axis accelerometer integrated into a wearable sixLowPAN device is utilised for this purpose and is in charge of gathering data collected from older people's movements in real-time. The Signals of the sensor are processed and analysed using a machine learning model on a sophisticated IoT gateway to give high efficiency in fall detection. We make use of low-cost wearable sensing devices from Apache Flink and MbientLab an open source broadcast engine, a short-term with a long memory network architecture, and categorization of fall. We examine the ideal Nyquist rate, sensor positioning, and multiple channeling information change using the training set, which was developed using the published dataset “MobiAct.” With a 95.87% accuracy rate, our edge computing system can detect falls using real-time data stream analytics.
Kulurkar et al. (Fri,) studied this question.