GGrantIndex
← Search

CI-SUSTAIN: Collaborative Research: Sustaining Successful Smartphone Testbeds to Enable Diverse Mobile Experiments

$366,000FY2016CSENSF

Regents Of The University Of Michigan - Ann Arbor, Ann Arbor MI

Investigators

Abstract

Mobile and wireless devices such as smartphones are transforming many aspects of our lives because of their computational and network capabilities. With the development of Internet of Things (IoT) based sensing and actuation support, wireless devices are also introducing a paradigm shift in various industry sectors such as advanced manufacturing, transportation, and personalized health. Despite this promise, it is still rather difficult to evaluate innovations in mobile computing science at scale due to limited access to realistic software and hardware platforms as well as user behavior. This award aims to support and sustain an already successful global-scale live laboratory known as MobiLab to further enable diverse mobile experiments in the long term. The proposed work will help augment both undergraduate lab-intensive introductory courses as well as graduate-level courses that expose students to various aspects in mobile computing. The resulting infrastructure will have a lasting impact on both the research community as well as industry by providing a rich repository of mobile system data from realistic use cases, software artifacts, and results based on the data. MobiLab will lighten the steep learning curve in mobile computing and IoT, as well as engage more researchers in this field. MobiLab has been conceived as a mobile computing analogue to the well-known planetary-scale networking testbed, PlanetLab, to support global-scale smartphone-based mobile computing science experimentation. Given many of the MobiLab open-source tools to support mobile experiments have already been widely deployed either in the form of crowd-sourced applications or as customized tools adopted in specific operational environment, this work focuses on developing several key software infrastructure improvements to enable long-term sustainability of the experimentation infrastructure. In particular, the project investigates how to complement the active and passive measurement support with a new programmable context-triggered measurement mechanism that helps balance the overhead of the measurement and the accuracy of the data captured. Another enhancement to the infrastructure is the design of metrics to accurately measure user quality of experience (QoE) based on models to be developed that allow QoE to be measured on unmodified devices and apps. QoE-based measurement support is integrated with the other MobiLab tools and deploy it at large scale via application marketplaces. Finally, this project also focuses on three main sustainability activities along the dimensions of community development, corporate sponsorship, as well as collaboration with other organizations and events.

View original record on NSF Award Search →