| The groundswell of wellness healthcare programs and patient management emphasize more involvement by patients themselves. This paradigm largely requires that patient information be readily available at the point of care. A great leap in the quality of medical care is within technological research if we can take advantage of new developments in sensor and computing technologies, wireless networking/communication, and biomedical informatics. Never before have there been tools available that bring the promise of reliable, convenient and continuous monitoring outside of the conventional clinical environment and potential for more individually tailored healthcare. Lightweight embedded systems are now gaining more popularity specially in medical systems due to the recent technological advances in fabrication that have resulted in more powerful tiny processors with greater communication capabilities that pose various scientific challenges for researchers. Reliability and lifetime of medical systems has been among the most important objectives in this field. While there has been years of research on these topics in computer and electrical engineering fields, lightweight embedded core of these devices pose new challenges.;This thesis encapsulates multiple medical system and application which we have developed and explores the research challenges in this area. Particularly, there is a focus on the most significant challenges which are energy consumption and reliability of these mobile and battery-powered systems. We will see why traditional reliability techniques fail in these systems and will present general methodologies for reliability-aware energy optimization for such systems with real-time constraints. Moreover, a new computational model will be presented which fits the unique characteristics of networked lightweight embedded systems. Under this model, network algorithms and problems are studied. Specially the fundamental problem of minimum spanning three is thoroughly investigated. |
