Abstract

In this study, we aim to develop a miniaturized stand-alone system that can detect a wide range of daily activities based on a single integrated consumer 3-axis accelerometer. A novel k-means based classification algorithm was constructed to interpret and translate signals from accelerometer into a recognizable cluster of pre-defined activities. The developed system has given encouraging results with a 100% success rate of classification of the three basic classes of activities based on resting, walking and running, and an 84% success rate for the lower level of different pace of walking and running. The potential extension towards self-monitoring systems for people suffering from diabetes mellitus has been considered by converting the activities into metabolic equivalents that will help predict the associated energy expenditure.