Application of statistical methods to improve an acceleration based algorithm

Abstract

Falls are the leading reason for death related accidents in people over 65 years old. Concerning this situation, it is necessary to develop a viable way of detecting these falls as fast as possible, so that medical assistance can be provided within useful time. In order for a system of this kind to work correctly, it must have a low percentage of false positives and a good autonomy. In this paper we present the research done in order to improve an existing acceleration based algorithm, which despite being inaccurate is however highly energy efficient. The study of its improvement was done resorting to the use of cluster analysis and logistic regression. The resulting algorithm distinguishes itself by being, at the same time, very accurate and having low energy consumption. © 2011 Springer-Verlag.