Power Saving Techniques for Wearable Devices

Approximately 85% of chronic disease patients demand healthcare services such as monitoring, diagnosis, treatment and prevention. Continuous physiological monitoring is key to providing preventative healthcare and accurate disease diagnosis, leading to a growing demand for autonomous wearable medical technology. Power optimizations are fundamental for such wearable medical devices because they are battery-powered and demand high power efficiency in order to extend their battery lifetime. Although there are multiple power management techniques, none of them is general enough to address the large variety of biomedical applications. In this work, we explore existing techniques of power reduction for wearable medical devices. Our analysis addresses software solutions to enable power reduction of wearable medical devices and their generalization for different medical signal processing applications. We propose a taxonomy of the power saving techniques. Task scheduling, compressed sensing or energy awareness are just some of the categories identified as common low-power techniques in wearable medical devices. The presented analysis helps to identify the most appropriate techniques based on the characteristics of the medical signal processing and to facilitate the generalization of those techniques, which are nowadays specific to medical applications.