Falls among elderly individuals are a serious global public health concern, causing significant injuries, disabilities, and fatalities. As aging bodies become weaker, the risk of accidental falls increases dramatically, especially for those with chronic illnesses or reduced mobility. Traditional fall alert systems often rely on manual activation, which is ineffective when the individual is unconscious or unable to respond. Recent advances in Internet of Things (IoT) technology, wearable sensor systems, and cloud-based alerting have enabled automatic detection of falls without user intervention. This survey paper consolidates research from three major IoT-based fall detection studies, analysing hardware components, sensing mechanisms, detection algorithms, and communication protocols. It highlights how accelerometers, gyroscopes, heart rate sensors, GPS, and cloud services can work together to provide accurate, real-time fall detection. The paper also discusses limitations such as false alarms, power consumption, and user acceptance, while proposing future improvements through Artificial Intelligence (AI), multimodal sensing, and healthcare ecosystem integration.
Thombare et al. (Fri,) studied this question.