11th International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services

Research Article

OSCAR: A Deployable Adaptive Mobile Bandwidth Sharing and Aggregation System

Download146 downloads
  • @INPROCEEDINGS{10.4108/icst.mobiquitous.2014.258056,
        author={Karim Habak and Khaled Harras and Moustafa Youssef},
        title={OSCAR: A Deployable Adaptive Mobile Bandwidth Sharing and Aggregation System},
        proceedings={11th International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services},
        publisher={ICST},
        proceedings_a={MOBIQUITOUS},
        year={2014},
        month={11},
        keywords={bandwidth aggregation collaborative bandwidth aggregation optimal bandwidth aggregation},
        doi={10.4108/icst.mobiquitous.2014.258056}
    }
    
  • Karim Habak
    Khaled Harras
    Moustafa Youssef
    Year: 2014
    OSCAR: A Deployable Adaptive Mobile Bandwidth Sharing and Aggregation System
    MOBIQUITOUS
    ICST
    DOI: 10.4108/icst.mobiquitous.2014.258056
Karim Habak1, Khaled Harras2, Moustafa Youssef3,*
  • 1: Georgia Tech
  • 2: CMU-Qatar
  • 3: E-JUST
*Contact email: moustafa.youssef@ejust.edu.eg

Abstract

The exponential increase in mobile data demand coupled with the rapid deployment of various wireless access technologies have led to the proliferation of multi-interface enabled devices. As a result, researchers focused on exploiting the available interfaces on such devices in both solitary and collaborative forms. Unfortunately, the proposed systems, that exploit these interfaces, face a formidable deployment barrier. Therefore, in this work, we present OSCAR, a system that exploits multiple network interfaces on modern mobile devices. OSCAR provides a set of mechanisms for sharing bandwidth across multiple collaborating devices and enables this collaboration by providing users with sharing incentives. We present an overview of the system coupled with the OSCAR scheduler. We evaluate OSCAR via implementation on Linux and compare our results to the current optimal achievable throughput. Our preliminary results shows that in the throughput maximization mode, we provide up to 150% enhancement compared to using only one interface, without any changes to legacy servers.