Physical Activity Classification and Monitoring using Artificial Neural Network

This paper provides an efficient way to design a physical activity classification and monitoring system using a wireless sensor network which consisting of cost sensitive tri-axial accelerometers. Physical activity increases the fitness level and exercise capacity of the human body and helps to reduce risk factors such as obesity, diabetes and extends the life expectancy. The main objective of this project is to develop a real-time and accurate physical activity monitoring system based on physical signal detection technique. To detect and classify multiple activities, the proposed system uses multi-sensor network which is able to overcome the limitations of a single accelerometer. It consists of an electronic device which is worn on the hip and finger of the person under test. The system can be used to monitor physiological parameters, such as temperature and physical activity of a human subject using temperature and accelerometer sensors. Artificial Neural Network is used to classifying the different physical activities such as jogging, cycling, normal and fast walking. Neural Network Toolbox in MATLAB is used to classify such kind of activities