Integrated wireless sensor system for efficient Pre-Fall detection
The life expectancy of humans in today’s era have increased to a very large extent due to the advancement of medical science and technology. The research in medical science has largely been focused towards developing methods and medicines to cure a patient after a diagnosis of an ailment. It is crucial to maintain the quality of life and health of the patient. It is of most importance to provide a healthy life to the elderly as this particular demographic is the most severely affected by health issues, which make them vulnerable to accidents, thus lowering their independence and quality of life. Due to the old age, most of the people become weak and inefficient in carrying their weight, this increases the probability of falling when moving around. This research of iterative nature focuses on developing a device which works as a preventive measure to reduce the damage due to a fall. The research critically evaluates the best approach for the design of the Pre-Fall detection system. In this work, we develop two wearable Pre-Fall detection system with reduced hardware and practical design. One which provides the capability of logging the data on an SD card in CSV format so that the data can be analyzed, and second, capability to connect to the Internet through Wifi. In this work, data from multiple accelerometers attached at different locations of the body are analyzed in Matlab to find the optimum number of sensors and the best suitable position on the body that gives the optimum result. In this work, a strict set of considerations are followed to develop a flexible, practical and robust prototype which can be augmented with different sensors without changing the fundamental design in order to further advance the research. The performance of the system to distinguish between fall and non-fall is improved by selecting and developing the most suitable way of calculating the body orientation. The different ways of calculating the orientation of the body are scrutinized and realized to compare the performance using the hardware. To reduce the number of false positives, the system considers the magnitude and the orientation to make a decision.
Rizkalla, Purdue University.
Computer Engineering|Electrical engineering
Off-Campus Purdue Users:
To access this dissertation, please log in to our