A Survey On Fall Detection Systems Using Wearable Devices

Falls represent a significant concern in healthcare, particularly for elderly individuals, as they often lead to severe injuries, hospitalizations, and fatalities. The growing prevalence of falls among older adults underscores the need for effective fall detection systems. Wearable devices equipped with sensors such as accelerometers and gyroscopes have emerged as promising solutions due to their affordability, portability, and ability to provide continuous monitoring. This survey systematically reviews existing fall detection systems utilizing wearable devices, categorizing them into threshold-based and machine learning-based approaches. Each technique is critically analyzed in terms of methodology, strengths, limitations, and performance metrics. Furthermore, this paper identifies key research gaps, including the need for adaptive sampling techniques, age-specific models, lightweight fusion algorithms for multi-sensor data integration, and hybrid approaches that combine threshold-based methods with machine learning. By synthesizing existing knowledge and highlighting future directions, this study aims to contribute to the development of more reliable, energy-efficient, and accessible fall detection technologies tailored for elderly care applications.