Energy-efficiency is a critical concern in continuously-running mobile applications, such as those for health and context monitoring. An attractive approach to saving energy in such applications is to defer the execution of delay-tolerant op- erations until a time when they would consume less energy. However, introducing delays to save power may have a detri- mental impact on the user experience. To address this prob- lem, we present Tempus, a new approach to managing the trade-off between energy savings and delay. Tempus saves power by enabling programmers to annotate power-hungry operations with states that specify when the operation can be executed to save energy. The impact of power manage- ment on timeliness is managed by associating delay bud- gets with objects that contain time-sensitive data. A static analysis and the run-time service ensure that power manage- ment policies will not delay an object more than its assigned budget. We demonstrate the expressive power of Tempus through a case study of optimizing two real-world applica- tions. Furthermore, laboratory experiments show that Tem- pus may effectively manage the energy-delay trade-off on re- alistic workloads. For example, in a news application, five Tempus annotations may be used to create a policy that re- duces the latency of downloading images 10 times compared to the original implementation without affecting energy con- sumption. Our experiments also indicate that the overhead of tracking budgets in Tempus is small.
While mobile sensing applications have broad application to domains such as intelligent transportation, environmental monitoring, and social networking, I am particularly interested in the application of this technology to the healthcare domain. In this talk, I will start by describing AudioSense – a novel mobile sensing application that allows audiologists to assess the performance of the hearing aids in the real-world. A key limitation of traditional laboratory and survey methods employed by audiologists is that they fail to predict when a hearing aid user will be dissatisfied with its performance in the real-world. In contrast with these techniques, AudioSense jointly characterizes both the user's auditory context and the performance of the hearing aid in that context. A field study involving 50 participants indicates that the contextual information is essential to identifying users that may be at risk of being dissatisfied with their hearing aid. In the second part of the talk, I will present some of the tools that we have created to simplify the development AudioSense and other mobile sensing applications. The focus is one of coordinating when different hardware resources (e.g., WiFi, 3G) are turned on and off to save energy without hindering user experience. I will present and a lightweight annotation language and middleware service that can be used to build energy-efficient mobile sensing applications for Android.
Octav Chipara is an Assistant Professor in the Department of Computer Science at the University of Iowa and part of the Aging Mind and Brain Initiative. He received his PhD from Washington University in St. Louis and completed his Postdoctoral Fellow at the University of California San Diego. My research focuses on the systems, networking, and software engineering aspects of developing mobile health (mHealth) systems that continuously monitor and infer the health status of patients in spite of operating in dynamic environments and on limited battery resources. The central theme of my research is that in order to harness the full potential of mHealth systems, we must have better tools for programming and analyzing their properties. My work combines the design of communication protocols, middleware, and programming tools with large-scale real-world deployments of working systems.