Learning, Prediction and Mediation of Context Uncertainty in Smart Pervasive Environments

摘要

The essence of pervasive computing lies in the creation of smart environments saturated with computing and communication capabilities, yet gracefully integrated with human users (inhabitants). Context Awareness is the most salient feature in such an intelligent computing paradigm. Examples of contexts include user mobility and activity among others. This paper reviews our work towards managing context uncertainty in smart pervasive environments. First we discuss a novel game theoretic learning and prediction framework that attempts to minimize the joint location uncertainty of inhabitants in multi-inhabitant smart homes. Next we present an ambiguous context mediation framework for smart home health care application. Finally, we describe an efficient, quality-of-inference aware context determination framework in pervasive care environments. We also present open problems in this area.