LAWRENCE — Glitches, delays or crashes on mobile devices can thwart attempts to watch the latest viral video, catch a big game or conduct a video chat. Not only are attempts to wirelessly access real-time events frequently plagued with problems, viewing such events seem to rapidly drain an inordinate amount of battery life from a smartphone or tablet.
Researchers at the University of Kansas School of Engineering are working to solve both issues in hopes of enabling more seamless video and data usage on a wireless device. Lingjia Liu, assistant professor of electrical engineering and computer science, is leading the effort and has secured a one-year, $122,000 grant from the National Science Foundation to begin work on the project. NSF will evaluate the progress on the research after the first year, then will determine whether to fund an additional two years of work.
“Through our research, we hope to discover how much of a power boost is actually needed for instantaneous data transmission, and we hope to find the optimum power level to ensure maximum energy efficiency that sends data without any delay,” Liu said.
The project centers on a level of data transmissions known as "delay-sensitive." Those are the snippets of communication that consumers expect to receive instantaneously, such as phone calls, video chats and live sporting events. Delay sensitive transmissions typically send data in the 20- to 50-millisecond range. Anything longer than that and users notice they’re not receiving information in real time. For communications that are delay-insensitive – such as web browsing or email – users typically accept a delay of a few seconds. Information is not expected instantaneously.
Liu said that improving energy efficiency for data transmission has been studied for years, but typically on delay-insensitive communications.
“We’re attacking an area where not much research has been done. There’s really no clear methodology to analyze this. That’s what we’re trying to develop,” Liu said. “It’s not an easy task, because power consumption on a smart phone is multifaceted. We’re focusing on one critical area that’s received little attention to this point.”
The project involves graduate students conducting research on a theoretical level and undergraduates working on the potential for implementing the technology.
“There’s a fundamental limit on energy efficiency – you can’t improve beyond a certain level. But we want to find algorithms that pinpoint the limit with the hope that eventually we can apply it on practical wireless devices,” Liu said.
Erik Perrins, professor of electrical engineering and computer science and a collaborator on the project, said smartphone users should be excited by the potential of the research, since it deals with two key issues: maximizing battery life and maximizing data rates.
“We are paying special attention to a certain kind of wireless traffic, two-way video, because right now wireless system designers are making educated guesses about how to allocate resources to deal with this traffic,” Perrins said. “As this kind of delay-sensitive traffic becomes more and more prevalent, solid design rules are needed, which is exactly what our research aims to deliver.”