Testbeds and Research Infrastructure: Development of Networks and Communities. 9th International ICST Conference, TridentCom 2014, Guangzhou, China, May 5-7, 2014, Revised Selected Papers

Research Article

Bandwidth Efficient Adaptive Live Streaming with Cooperative Devices in Mobile Cloud Computing

Download90 downloads
  • @INPROCEEDINGS{10.1007/978-3-319-13326-3_28,
        author={Xiaoyi Zhang and Geng Xi and Kaiming Qu and Lin Zhang},
        title={Bandwidth Efficient Adaptive Live Streaming with Cooperative Devices in Mobile Cloud Computing},
        proceedings={Testbeds and Research Infrastructure: Development of Networks and Communities. 9th International ICST Conference, TridentCom 2014, Guangzhou, China, May 5-7, 2014, Revised Selected Papers},
        proceedings_a={TRIDENTCOM},
        year={2014},
        month={11},
        keywords={Dynamic Adaptive Streaming Mobile Cloud Computing Cooperative Smartphone},
        doi={10.1007/978-3-319-13326-3_28}
    }
    
  • Xiaoyi Zhang
    Geng Xi
    Kaiming Qu
    Lin Zhang
    Year: 2014
    Bandwidth Efficient Adaptive Live Streaming with Cooperative Devices in Mobile Cloud Computing
    TRIDENTCOM
    Springer
    DOI: 10.1007/978-3-319-13326-3_28
Xiaoyi Zhang1,*, Geng Xi1,*, Kaiming Qu1,*, Lin Zhang1,*
  • 1: Beijing University of Posts and Telecommunications
*Contact email: Zhangxy_bupt@126.com, as_to@msn.cn, qukm90@gmail.com, zhanglin@bupt.edu.cn

Abstract

Recent developments have heightened the optimization of Dynamic Adaptive Streaming over HTTP (DASH) services in mobile network condition (e.g. LTE/WIFI networks). The aim of this paper is to discuss how Mobile Cloud Computing (MCC) can assist DASH in a special scenario that a set of neighboring mobile devices take interests in watching the identical video stream. A mechanism is proposed to provide higher resolution live streaming with less expense (both in bandwidth and dollar-cost per device) by the cooperation among devices. The cloud-based live stream server will transcode the original stream segment according to the estimation of the devices’ group bandwidth, every device then share received fragments (part of the segment) with each other through the free device-to-device interface, and finally gets the whole segment. An emulation testbed is realized with Android Smartphone implementation according to the proposed improvement to DASH. Experiments results demonstrate its performance with today’s commercial players on the Quality of Experience (QoE), Peak Signal to Noise Ratio (PSNR) and network utilization across a range of scenarios.