• DocumentCode
    3308330
  • Title

    Optimization of non-cooperative P2P network from the game theory point of view

  • Author

    Rozario, Fabian ; Han, Zhu ; Niyato, Dusit

  • Author_Institution
    Dept. of Electr. & Comput. Eng. Dept., Univ. of Houston, Houston, TX, USA
  • fYear
    2011
  • fDate
    28-31 March 2011
  • Firstpage
    868
  • Lastpage
    873
  • Abstract
    BitTorrent is a peer-to-peer (P2P) file sharing protocol, which overcomes the bottleneck of server bandwidth by allowing peers to share the downloaded data directly. However, the greedy and non-cooperative behavior (such as leeching) of individual peer can degrade the performance of P2P significantly. In this paper, we optimize the non-cooperative P2P network from the game theory point of view. Specifically, we investigate the scenarios of pure Nash equilibrium, mixed Nash equilibrium, correlated equilibrium, and pure cooperation. We study the users´ behaviors as a good peer (i.e., uploading data to others) or a bad peer (i.e., only downloading data). The results provide a guidance for the server with the presence of greedy peers. Analysis and simulation results show the network behaviors under the different scenarios, which can provide guidance to practical distributed design in large scale networks.
  • Keywords
    access protocols; game theory; optimisation; peer-to-peer computing; BitTorrent; correlated equilibrium; game theory; greedy peers; leeching; mixed Nash equilibrium; noncooperative P2P network; optimization; peer-to-peer file sharing protocol; pure cooperation; server bandwidth; Delay; Games; Nash equilibrium; Peer to peer computing; Servers; Simulation; Correlated Equilibrium; Nash Equilibrium; P2P Network;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Wireless Communications and Networking Conference (WCNC), 2011 IEEE
  • Conference_Location
    Cancun, Quintana Roo
  • ISSN
    1525-3511
  • Print_ISBN
    978-1-61284-255-4
  • Type

    conf

  • DOI
    10.1109/WCNC.2011.5779246
  • Filename
    5779246