Estimation of the Optimal Radar Positioning for the Contactless Fall Detection

To the date, the elderly fall episodes pose a serious threat to their health and life, thus, the development of the effective and precision method for the fall detection remains one of the most relevant problem for the geriatry. Bioradiolocation-based methods have proven their effectiveness in automatic contactless detection of the elderly people falls in the recent studies, though, some aspects of the method are still to be researched. In particular, in the present paper we estimate the influence of the different radar placement height on the reliability of high-intense movement artifacts detection. A series of experiments with different types of movements were conducted and an experimental database of bioradar signals has been built. Then, a number of signal features have been calculated to discover the difference between bioradar signals with no movement and with the movement artifacts detected. The most informative features have been chosen for the non-parametric statistical tests to check the statistical significance of the results and to draw a conclusion about the preferable radar placement height. The results show that lower height appears to be more effective in the movement artifacts detection task.