Award Abstract # 1250180
CSR: EAGER: Network Traffic Aware Smartphone Energy Savings

NSF Org: CNS
Division Of Computer and Network Systems
Recipient: COLLEGE OF WILLIAM AND MARY
Initial Amendment Date: August 25, 2012
Latest Amendment Date: August 25, 2012
Award Number: 1250180
Award Instrument: Standard Grant
Program Manager: Marilyn McClure
mmcclure@nsf.gov
 (703)292-5197
CNS
 Division Of Computer and Network Systems
CSE
 Directorate for Computer and Information Science and Engineering
Start Date: September 1, 2012
End Date: August 31, 2015 (Estimated)
Total Intended Award Amount: $200,000.00
Total Awarded Amount to Date: $200,000.00
Funds Obligated to Date: FY 2012 = $200,000.00
History of Investigator:
  • Gang Zhou (Principal Investigator)
    gzhou@wm.edu
Recipient Sponsored Research Office: College of William and Mary
1314 S MOUNT VERNON AVE
WILLIAMSBURG
VA  US  23185
(757)221-3965
Sponsor Congressional District: 08
Primary Place of Performance: College of William and Mary
Office of Sponsored Programs
Williamsburg
VA  US  23187-8795
Primary Place of Performance
Congressional District:
01
Unique Entity Identifier (UEI): EVWJPCY6AD97
Parent UEI: EVWJPCY6AD97
NSF Program(s): CSR-Computer Systems Research
Primary Program Source: 01001213DB NSF RESEARCH & RELATED ACTIVIT
Program Reference Code(s): 7916
Program Element Code(s): 735400
Award Agency Code: 4900
Fund Agency Code: 4900
Assistance Listing Number(s): 47.070

ABSTRACT

Modern smartphones are notorious for consuming energy far too quickly, especially for access to the Internet through wireless radios.
Although certain advances have been made on the hardware side such as better batteries, this research is focused on improving energy management software to make better use of existing batteries.
Considering that the frequency of use of these radios is spurred by the popularity of smartphone applications, the wide availability of applications, for instance the Android Market has over 500,000 registered applications, shows that the need to save smartphone radio energy is highly relevant and urgent today.

Different from existing research, this high-risk high-reward research takes a network traffic aware approach to save smartphone radio energy. The main intellectual merits include: (1) Solutions for determining low priority delay tolerant smartphone applications without assistant from application developers. (2) Solutions for determining delay tolerant traffic periods within high priority real-time smartphone applications. (3) Solutions for tracking low priority delay tolerant applications, and delay tolerant traffic periods in high priority real-time applications through the system to optimize radio energy efficiency.

This research will extend smartphone battery lifetime with the potential to benefit over one billion smartphone users worldwide. User experience of smartphone usage will be enhanced. Industry smartphone practice will also be improved. This research will be integrated into three courses, and also benefit research of female graduate students and undergraduate students. Special distribution efforts are also planned to increase the number of women and girls in computing.

PUBLICATIONS PRODUCED AS A RESULT OF THIS RESEARCH

Note:  When clicking on a Digital Object Identifier (DOI) number, you will be taken to an external site maintained by the publisher. Some full text articles may not yet be available without a charge during the embargo (administrative interval).

Some links on this page may take you to non-federal websites. Their policies may differ from this site.

Andrew Pyles, David Nguyen, Xin Qi, Gang Zhou "Bluesaver: A Multi PHY Approach to Smartphone Energy Savings" IEEE Transactions on Wireless Communications , 2015
Conner kasten, Gang Zhou "Typed VoIP Silence Prediction for Smartphone Energy Savings" [Springer WPC?14] Springer Wireless Personal Communications , 2014

PROJECT OUTCOMES REPORT

Disclaimer

This Project Outcomes Report for the General Public is displayed verbatim as submitted by the Principal Investigator (PI) for this award. Any opinions, findings, and conclusions or recommendations expressed in this Report are those of the PI and do not necessarily reflect the views of the National Science Foundation; NSF has not approved or endorsed its content.

This project aims to save smartphone energy with a network traffic aware approach. We start this project by research on two main aspects: exploiting application priority for energy savings, and exploiting delay tolerant time periods within high priority, i.e. delay-sensitive, applications for energy savings.

First, we invent a set of novel solutions that prioritize different applications, in terms of their delay tolerance, based on network traffic analysis, and then use it to optimize smartphone radio energy. For example, our SAPSM solution is able to learn a smartphone user’s delay-tolerance, in terms of whether she/he can tolerant 100ms~200ms delay for specified applications, through a network traffic analysis based machine learning. This solution demonstrates up to 56% energy saving in an Android phone implementation.  

Second, for applications that, in general, are considered delay-sensitive, including VoIP and live video streaming, we develop a set of novel solutions that identify delay-tolerant segments within delay-sensitive applications, and then during those segments put radio to low power mode to save energy. For example, our SiFi solution is able to put WiFi in low power mode while nobody is speaking through VoIP, achieving 40% energy saving for an Android smartphone.

Third, we research on several cross-cutting issues of our network traffic aware approach. We investigate the impact of daily WLAN broadcast traffic on smartphone energy efficiency, and devise a family of strategies to minimize energy consumption at the presence of dynamic broadcast traffic; We develop solutions that prioritize and terminate smartphone background traffic for energy saving; We analyze and compare traffic-aware and non-traffic features in prioritizing smartphone applications; We build typology of silences for VoIP traffic and compare it with un-typed SiFi solution; We utilize sensor traffic on smartphone to assist network traffic analysis; We also invent a hybrid Bluetooth/WiFi approach for network traffic optimization.

As a broader impact, our research on network traffic analysis enables us to apply a similar approach on storage traffic analysis for smartphone energy saving. We also optimize smartphone delay while saving energy. As another broader impact, we incorporate our smartphone energy saving solutions in smartphone-centered body sensor networks for healthcare applications.

In total, 17 academic papers have been published based on research results from this NSF project. Four PhD students (one female), 3 Master students (one female), two undergraduate students (one female), and one high school student have contributed to this project. Three U.S. patents have been filed (one awarded, two pending) based on our research results. Commercializing efforts are still ongoing.  


Last Modified: 09/25/2015
Modified by: Gang Zhou