This work took place in the development of a remote home healthcare monitoring application designed to detect distress situations through several types of sensors. The multi-sensor fusion can provide more accurate and reliable information compared to information provided by each sensor separately. Furthermore, data from multiple heterogeneous sensors present in the remote home healthcare monitoring systems have different degrees of imperfection and trust. Among the multi-sensor fusion techniques, belief methods based on Dempster-Shafer Theory are currently considered as the most appropriate for the representation and processing of imperfect information, thus allowing a more realistic modeling of the problem. Based on a graphical representation of the Dempster-Shafer called Evidential Networks, a structure of heterogeneous data fusion from multiple sensors for fall detection has been proposed in order to maximize the performance of automatic fall detection and thus make the system more reliable. Sensors’ non-stationary signals of the considered system may lead to degradation of the experimental conditions and make Evidential Networks inconsistent in their decisions. In order to compensate the sensors signals non-stationarity effects, the time evolution is taken into account by introducing the Dynamic Evidential Networks which was evaluated by the simulated fall scenarios corresponding to various use cases
