The rapid growth of non-invasive sensing and low-power wireless communication technologies has enabled continuous vital sign acquisition using wearable biomedical sensing and processing motes. These sensor motes are capable of acquiring vital signs such as blood pressure and flow, temperature, Electro CardioGram (ECG), oxygen saturation, and CO2 concentration. However, simultaneously executing compute-intensive models for deriving physiological parameters from these vital signs and for acquiring context awareness in real time requires computing capabilities that go beyond those of an individual sensor mote's and/or portable device's.
The objective of the proposed research is to enable real-time in-situ vital sign data processing so to extract non-measurable physiological parameters, to interpret it under context, and to acquire actionable knowledge about a person's health. To realize this objective, the collective computational capabilities of laptops, tablets, smartphones, and desktop computers in the vicinity as well as remote storage and compute servers need to be exploited. This research project will focus on the fundamental research challenges to organizing these resources into an elastic resource pool (a hybrid computing grid). The most significant challenge is presented by the inherent uncertainty in the mobile grid environment that can be attributed to unpredictable node mobility, varying rate of battery drain, and high susceptibility to hardware failures. The significant contributions of the proposed research are i) a role-based architectural framework for reliable grid coordination under uncertainty, i.e., for handling resource/service discovery, service request arrivals, and workload distribution and management, and ii) a novel uncertainty- and energy-aware resource allocation engine, which will distribute the workload tasks optimally among the networked computing devices so to ensure Quality of Service (QoS) in terms of application response time and energy consumption.