Abstract
In this paper, we introduce a Health signs and Activity recognition MONitoring framework (Hamon). Hamon, of German origin meaning a home protector [13], is designed to be an enabling prototype for health monitoring applications. As one of the possible applications, we implemented an activity detection prototype using off-the-shelf sensors. The new activity recognition algorithm we present here is based on accelerometers signals, a K-nearest neighbor (KNN) classifier and a Bayesian network. The framework collects and analyzes sensory data in real-time, and provides different feedback to the users. In addition, it can generate alerts based on the detected events and store the data collected to a medical sever. Context information such as the weather condition, the type of activity, and physiological data collected such as the heart rate, is also integrated in the framework.