Want to Play DASH?

A Game Theoretic Approach for Adaptive Streaming over HTTP



Abdelhak Bentaleb*     Ali C. Begen§     Saad Harous‡     Roger Zimmermann*    
*National University of Singapore, §Ozyegin University, ‡United Arab Emirates University

  • Abstract

    In streaming media, it is imperative to deliver a good viewer experience to preserve customer loyalty. Prior research shows that this is rather diffcult when shared Internet resources struggle to meet the demand from streaming clients that are largely designed to behave in their own self-interest. To date, several schemes for adaptive streaming have been proposed to address this challenge with varying success. In this paper, we take a different approach and develop a game theoretic approach. We present a practical implementation integrated in the dash.js reference player and provide substantial comparisons against the state-of-the-art methods using trace-driven and real-world experiments. Our approach outperforms its competitors in the average viewer experience by 38.5%, and in video stability by 62%.

    • Demo


    Offline Demonstration

    We consider two main scenarios: one player and multiple players. We use this demo to simplify the emulation of the obtained results in GTA paper via an interactive, easy, effective and useful demonstration. The users are given the possibility to select the test scenario, different content types, throughput profile and the ABR schemes to compare against each other. Moreover, for each ABR scheme, the viewer QoE metrics of the video session are reported in the form of blue text, and also it shows the bitrate and its variability in a visible graph.

    Tips for the offline demo

    Online Demonstration

    The online demo emulates a realistic scenario in real time. This demo gives the user the ability to play around various operating regimes and settings. They can test and add whatever they want and build a customized solution.

    Tips for the online demo



    Videos



    Offline Video 1
    Offline Video 2
    Offline Video 3
    Online Video 1
    Online Video 2


    Papers

    A Distributed Approach for Bitrate Selection in HTTP Adaptive Streaming
    Abdelhak Bentaleb, Ali C. Begen, Saad Harous and Roger Zimmermann
    ACM Multimedia 2018 [pdf] [ACM DL]

    Want to Play DASH? A Game Theoretic Approach for Adaptive Streaming over HTTP
    Abdelhak Bentaleb, Ali C. Begen, Saad Harous and Roger Zimmermann
    ACM MMSys 2018 [pdf] [ACM DL]
     Best Student-paper Award
    DASH-IF DASH-IF Excellence in DASH Award (1st place)

    Game Theory Based Bitrate Adaptation for dash.js Reference Player
    Abdelhak Bentaleb, Ali C. Begen and Roger Zimmermann
    IEEE ICME 2018 [pdf] [IEEExplore]
    DASH-IF Grand Challenge on DASH (1st place)


    Code

    The instructions for testing and adapting the code are given in the Appendix Section of our MMSys'18 paper.

    Quick Setup

    Detailed Setup



    © 2018 Streaming University

    Contact Us





    Offline Video 1

    Scenario:1 Player Demo.
    Throughput variability trace:DASH-TH4.
    Content type:Animation.
    ABR scheme:All.



    Offline Video 2

    Scenario:1 Player Demo.
    Throughput variability trace:3G/HSDPA.
    Content type:Sport.
    ABR scheme:All.



    Offline Video 3

    Scenario:100 Player Demo.
    Content type:Animation.
    ABR scheme:All.



    Online Video 1

    Scenario:1 Player Demo.
    Content type:Animation.
    ABR scheme:GTA, Hybrid.



    Online Video 2

    Scenario:1 Player Demo.
    Content type:Documentary.
    ABR scheme:GTA, Hybrid.



    Tips for the offline demo

    Step 1: Select the scenario: 1 Player Demo or 100 Players Demo.
    Step 2: Select the throughput variability trace: FCC, 3G/HSDPA, Synthetic, DASH-[TH2,TH3,TH4].
    Step 3: Select the content type: Animation, Movie, Documentary, Sport.
    Step 4-5: Select the ABR scheme: GTA, Rate-based, Buffer-based, Hybrid.
    Step 6: Click on "Load" to start the video streaming sessions.
    Step 7: The reported QoE metrics: Bitrate, Avg.Bitrate, Bitrate Switch, Stall Time.


    Tips for the online demo

    Step 1: Click on "Stream" to select the content type.
    Step 2: Select the content type: Animation, Movie, Documentary, Sport.
    Step 3: Click on "Load" to start the video streaming session.
    Step 4: The reported QoE metrics.