A Prototype for Resource Optimized Context Determination in Pervasive Care Environments

Abstract

In this demonstration we demonstrate an early prototype that shows the importance of context determination in the presence of mobile entities in pervasive care environments. We use our novel context model to build a framework for resource-constrained sensor networks. We then use this context model to use a user’s mobility to infer his activity, which we refer to as his context state. Because the context state is inferred from actual sensed context, we use the prototype to demonstrate the tradeoff between context inferencing accuracy and communication overhead. The ability to sense the environment and accurately infer context can help monitor the user in a pervasive care environment.