Applicability of Mel-Cepstrum-Related Features for Action Recognition Based on Data from Impulse-Radar Sensors

This paper is devoted to research on the application of impulse-radar technology for unobtrusive monitoring of elderly persons. The applicability of mel-cepstrum-related features for action recognition by an impulse-radar-based monitoring system is explored. The data from the impulse-radar sensors are first transformed into the monitored person's three-dimensional coordinates which are then used to calculate features characterising that person's movement. The findings from the real-world experiments show that the proposed set of features can be used to improve the accuracy of recognition of actions such as walking, sitting and lying down.