Server-assisted context-dependent pervasive wellness monitoring

Recent research on remote health monitoring solutions has focused largely on developing context-dependent, streamprocessing capabilities on a personal mobile hub (typically, a cellphone) for energy-efficient transmission of data collected from a set of body-worn medical sensors. In this paper, we argue that commercial deployment of such pervasive wellness monitoring will require the extension of such `context dependency' to the process of data collection (from the sensors to the mobile device) as well. In particular, the utilization of an individuals non-medical activity context, by the cellphone, in coordination with a backend server, is posited to be the key to supporting important objectives such as intermittent sensing and data security/privacy. The server-side platform is able to extract relevant context from a wide variety of personal or generic backend information streams; this context is then enriched by local context derived by the smartphone. We present an early outline of a middleware platform for supporting this objective.