HCC: Human Factors in Energy-Efficient Mobile Computing System Design

William Marsh Rice University, Houston TX

Investigators

Abstract

The goal of this research project is to systematically address the human factors in energy-efficient design of mobile computing systems. The project consists of three components. First, it will study how mobile users deal with and adapt to limited battery lifetime. It will subsequently apply the new knowledge to the design of system algorithms and their user interfaces for users to manually make effective tradeoffs between battery and other aspects of system usability. Second, this project will investigate the user impact of power-saving mechanisms and its context dependence. It will employ the Orbit body-area sensor nodes and mobile phone-based in situ survey tools to collect context information and subsequently build models of the context-dependent user impact. Using these models, the research project will employ information from user interfaces and body-area sensors to automatically apply power-saving mechanisms and adapt the system in a user-friendly way. Third, this project will explore tools and methodologies for designing energy-efficient graphical user interfaces on emerging OLED and bistable display technologies, which have dramatically different power characteristics from the conventional liquid crystal displays. The research project will employ both laboratory and field studies with technologies from quantitative self-reporting, qualitative ethnographic inquiries, and computer engineering. Its user studies and field trials will involve mobile users from under-served urban communities. Mobile devices, in particular mobile phones, have shown great potential to provide low-income people with usable and affordable access to information technologies. Their limited battery lifetime and limited heat dissipation capability have become a critical challenge to achieving such potential. This research project has the potential to provide enabling technologies and tools that address this challenge thus help bridge the digital divide nationally and globally. It will make fundamental contributions to human-centered design of mobile computing systems as well as provide novel user interfaces, methodologies, algorithms, and their tool implementations for better battery usability and longer battery lifetime. The research endeavor will bring foundational knowledge into both human-centered computing and low-power design as well as bring new opportunities into mobile and ubiquitous computing. It is likely to have a broader impact in the industry through the PI's collaborations with industrial leaders. The research results will be incorporated into undergraduate and graduate courses in mobile computing. The tools, collected data, and educational content will be made open-source and on-line.