Award Abstract # 1812399
NeTS: Small: Collaborative Research: Network-Centric Mobile Cloud Computing

NSF Org: CNS
Division Of Computer and Network Systems
Recipient: UNIVERSITY OF PITTSBURGH - OF THE COMMONWEALTH SYSTEM OF HIGHER EDUCATION
Initial Amendment Date: December 15, 2017
Latest Amendment Date: December 15, 2017
Award Number: 1812399
Award Instrument: Standard Grant
Program Manager: Murat Torlak
CNS
 Division Of Computer and Network Systems
CSE
 Directorate for Computer and Information Science and Engineering
Start Date: September 1, 2017
End Date: September 30, 2020 (Estimated)
Total Intended Award Amount: $204,690.00
Total Awarded Amount to Date: $204,690.00
Funds Obligated to Date: FY 2015 = $204,690.00
History of Investigator:
  • Wei Gao (Principal Investigator)
    weigao@pitt.edu
Recipient Sponsored Research Office: University of Pittsburgh
4200 FIFTH AVENUE
PITTSBURGH
PA  US  15260-0001
(412)624-7400
Sponsor Congressional District: 12
Primary Place of Performance: University of Pittsburgh
PA  US  15213-2303
Primary Place of Performance
Congressional District:
12
Unique Entity Identifier (UEI): MKAGLD59JRL1
Parent UEI:
NSF Program(s): Networking Technology and Syst
Primary Program Source: 01001516DB NSF RESEARCH & RELATED ACTIVIT
Program Reference Code(s): 7923, 9150
Program Element Code(s): 736300
Award Agency Code: 4900
Fund Agency Code: 4900
Assistance Listing Number(s): 47.070

ABSTRACT

Mobile cloud computing (MCC) has been used to address the resource limitation of mobile devices by migrating expensive local computations to the cloud. However, transmitting data wirelessly from mobile devices to the cloud also consumes energy. Hence, the key problem of MCC is how to minimize the energy consumption while preserving the mobile application performance. Different from traditional solutions which focus on reducing the cost of wireless transmission solely from the application perspective, this project focuses on designing MCC schemes from a network-centric perspective, by investigating, formulating, and mitigating the impact of special characteristics of wireless networks on the energy efficiency of MCC. The proposed research could benefit end users with various mobile devices by extending their battery lifetime and improving their performance. The results from this research are likely to foster new research directions on supporting MCC from a network-centric perspective. The project will engage under-represented students in the proposed research, and the scholarly discovery of this project will be disseminated broadly to the community.

This project aims to improve the performance of MCC by mitigating the impacts of two special characteristics of wireless networks: the long-tail problem at the wireless interface and the quality variations of the wireless link. More specifically, this project consists of three closely intertwined research thrusts: (i) reducing the amount of tail energy when transmitting the program states to the remote cloud, while ensuring that the performance requirements of mobile applications can be met; (ii) mitigating the impact of wireless link quality on both energy and performance, and minimizing the degradation of application performance when the wireless link quality is low; and (iii) exploiting the difference of wireless link quality among mobile users to further improve the energy efficiency of MCC via user cooperation. An experimental testbed will be developed to investigate the practical impact of wireless network characteristics on MCC and evaluate the proposed MCC schemes.

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.

Haoyang Lu and Wei Gao "Continuous Wireless Link Rates for Internet of Things" Proceedings of the 17th ACM/IEEE International Conference on Information Processing in Sensor Networks (IPSN) , 2018
Haoyang Lu, Ruirong Chen and Wei Gao "EasyPass: Combating IoT Delays with Multiple Access Wireless Side Channels" in Proceedings of the 15th International Conference on emerging Networking EXperiments and Technologies (CoNEXT) , 2019
Ruirong Chen and Wei Gao "Enabling Cross-Technology Coexistence for Extremely Weak Wireless Devices" IEEE INFOCOM , 2019
Ruirong Chen and Wei Gao "StarLego: Enabling Custom Physical-Layer Wireless over Commodity Devices" in Proceedings of the 21st International Workshop on Mobile Computing Systems and Applications (HotMobile) , 2020
Yihao Liu, Kai Huang, Xingzhe Song, Boyuan Yang and Wei Gao "MagHacker: Eavesdropping on Stylus Pen Writing via Magnetic Sensing from Commodity Mobile Devices" in Proceedings of the 18th ACM International Conference on Mobile Systems, Applications, and Services (MobiSys) , 2020
Yong Li and Wei Gao "Minimizing Context Migration in Mobile Code Offload" IEEE Transactions on Mobile Computing , v.16 , 2017
Yong Li and Wei Gao "MUVR: Supporting Multi-User Mobile Virtual Reality with Resource Constrained Edge Cloud" ACM/IEEE Symposium on Edge Computing , 2018

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.

One of the most important challenges in mobile computing is how to prolong the limited battery life of mobile devices when running computationally expensive mobile applications. These applications either execute complicated algorithms to recognize humans’ voice and physical gestures, or involve expensive multimedia computations for gaming and video playback. A viable solution to this challenge is mobile cloud computing (MCC), which migrates local computation to the remote cloud via wireless networks. Such remote program execution is particularly important to the recently emerged wearable devices such as smart watches.

To support remote program execution in MCC, the program states are wirelessly transmitted between mobile devices and the cloud. Wireless transmission consumes a large amount of energy and determines the energy efficiency of MCC. The key problem of MCC, hence, is how to minimize the energy while maintaining the mobile application performance. To deal with this problem, researchers suggested to adaptively partition a mobile application and only migrate the portion that benefits the most for remote execution to the cloud. These traditional solutions, however, have limitations in that they reduce the cost of wireless transmission solely from the application perspective, by reducing the size of program states being transmitted according to the network bandwidth and programs’ computational complexity They have minimal explorations into the special characteristics of wireless networks, which affect the energy consumption of wireless transmission and have a pivotal role in energy-efficient MCC. Ignorance of these wireless network characteristics is also the major factor hindering practical integration of mobile devices into the cloud.

To address these challenges, we propose to design MCC schemes from a network-centric perspective, by investigating, formulating, and mitigating the impact of special characteristics of wireless networks on the energy efficiency of remote program execution in MCC. We will incorporate these characteristics to ensure that each program state is transmitted to the cloud in the most energy-efficient manner.

First, we designed wireless networking protocols and systems to exploit the unique wireless networking characteristics for more efficient MCC operations. We have built a wireless side channel that could effectively discover and utilize the available SNR margin in wireless networks to maximize the efficiency of workload migration between mobile devices and the cloud. We have developed application-aware traffic scheduling protocols for better workload offloading in mobile clouds.

Second, we have developed mobile system techniques that support code offload with the least amount of context migration in the mobile cloud. Based on this technique, we further proposed and realized a resource sharing framework that allows multiple mobile devices in a personal mobile cloud to efficiently and seamlessly exchange and share their computing and sensing resources with each other. This framework is highly adaptive and able to autonomously adapt to heterogeneous wireless network dynamics and changes.

Third, we have been applying these system techniques and designs to a large variety of MCC applications. In particular, we used these techniques to facilitate MCC-assisted mobile Virtual Reality, where the mobile head-mounted displays (HMDs) offload the expensive VR computing workloads to the cloud servers and received the generated VR contents accordingly. Adaptive computing workload migration and VR contents transmissions are being conducted between these mobile devices and cloud servers, to make sure that the contingent VR application performance requirements can be always met.

We have implemented the above system designs over practical mobile system testbeds, and conducted various real experiments to demonstrate their applicability in real-world mobile cloud scenarios. Such experimentation includes code instrumentation over a large collection of mobile applications, and demonstrated significant performance improvement over different mobile platforms.

Many of the research results have been integrated with the education curricula at the University of Tennessee Knoxville and the University of Pittsburgh. The project has supported multiple Ph.D.  students working on their dissertations, and the research outcome has also been used to involve undergraduate students into research. The involvement of the graduate and undergraduate students will prepare them for leadership roles in computer science research, academia, and industry. 


Last Modified: 01/01/2021
Modified by: Wei Gao

Please report errors in award information by writing to: awardsearch@nsf.gov.

Print this page

Back to Top of page