Rethinking Context for Pervasive Computing: Adaptive Shared Perspectives

Mobile and pervasive computing applications depend on environmental awareness, not just of their own local system and immediate surroundings, but increasingly of the network as a whole. Traditional models of context do not sufficiently account for the trade-off between accuracy, availability, and efficiency when considering the mechanisms by which context can be gathered and shared in mobile, ad-hoc networks. Traditional approaches to context remain egocentric, but next generation environments demand coordination among entities and access to shared context resources, and thus context must be treated as a combination of local and shared information. We motivate our perspective on the future of context in mobile pervasive environments staring with historical framing of the problem, and present our recent efforts in adaptive shared context perspectives.