Release Date: October 10, 2014
BUFFALO, N.Y. — Charger-toting smartphone users have reason to rejoice.
Three University at Buffalo research groups have received $1.3 million in combined funding to improve energy management in smartphones.
Each team attacks the issue from a different perspective: hardware, operating systems and networking. The projects range from determining which phone functions and hardware sap the most energy, to developing a mobile application that allows users to prioritize how much power each app can use.
The grants were awarded by the National Science Foundation’s Computer and Network System (CNS) Division, which supports research and education that lead to the invention of new technologies or the improvement of existing equipment.
Prioritizing energy with power agility: CNS Medium Program ($600,000 grant divided with Drexel University)
It would make little sense for a company to give two delivery drivers the same amount of gas if one of the drivers had to travel twice as far.
Now, imagine if the company was your phone and the drivers its applications. Simple and power-intensive apps are given the same amount of energy to use, deteriorating the user experience with the larger app — familiar to most people recognize as buffering.
“Power agility,” a project led by Geoffrey Challen, PhD, assistant professor in the Department of Computer Science and Engineering, is a system that allows users to prioritize the amount of energy a smartphone provides its applications.
“For a long time, operating systems have been given this task of determining what the allocations should be, and that is not the right way,” says computer science and engineering doctoral candidate Guru Prasad, who is developing power agility under Challen. “Operating systems have no idea what applications are doing or what each app needs.”
Challen, with the assistance of Prasad and computer science and engineering graduate student Scott Haseley, are gauging power allocation based on app efficiency.
The most efficient rate at which a phone can operate is the slowest, but this comes at the cost of the user’s experience. The ideal inefficiency is the slowest speed at which an app can operate while maintaining a positive user experience.
Power agility attempts to compute the optimal inefficiency of smartphone apps and then provides the data to operating systems to assign energy-distribution priorities. With the UB system, users can prioritize apps based on importance.
Researchers at Drexel University are partnering with UB engineers on the study.
Power and performance tradeoffs in next generation WiFi: CNS Small Program ($500,000 grant)
Most smartphone-technology advancements come at a cost to battery life. WiFi is no exception. Connecting to networks — WiFi, 4G, etc. — can account for up to 50 percent of a phone’s power budget.
Dimitrios Koutsonikolas, PhD, assistant professor, and Ramalingam Sridhar, PhD, associate professor, both in the Department of Computer Science and Engineering, are leading a study to measure power consumption in smartphones and develop models that will lead to new, efficient WiFi protocols.
Using several smartphone devices, the researchers will measure energy consumption across several WiFi configurations and applications, along with power used by hardware such as the processor and screen. The group also will sample users to determine average behavior surrounding energy consumption.
With this three-pronged approach, researchers can develop models that capture the tradeoffs between power consumption and performance, and describe the impact of different features and behavior on WiFi-capable devices.
The models will guide the design of new WiFi protocols that will allow the next generation of smartphones, tablets and laptops to experience faster data speeds without exhausting battery life.
Swetank Kumar Saha, a computer science and engineering doctoral candidate, is involved in the study.
Jouler, an energy management app: CNS Small Program ($500,000 grant)
Because people vary in how they use their smartphones, it’s impossible to take a one-size-fits-all approach to prioritizing energy distribution.
“Right now, your Android phone is trying to manage energy in some way; it’s just not doing a very good job for a variety of users,” says Challen.
Challen is supervising Anudipa Maitii, a computer science and engineering doctoral candidate, on a project that involves a new category of apps called energy managers.
Their creation, Jouler — a play on the name for a unit of energy — is an Android app for Google Play that manages battery life for smartphone users. Its abilities include delaying tasks such as synching, changing the bandwidth of other applications and prioritizing the amount of energy apps are allowed to use.
In order to grant Jouler these permissions, researchers need to alter a phone’s operating system, which restricts the application to Android platforms.
“Jouler exposes energy-management controls hidden in the operating system, finally giving the user a choice on preserving their phone’s battery,” says Maitii.