• DocumentCode
    1684322
  • Title

    Network embedded FEC (NEF) performance over multi-hop wireless channels with memory

  • Author

    Wu, Mingquan ; Radha, Hayder

  • Author_Institution
    Dept. of Electr. & Comput. Eng., Michigan State Univ., East Lansing, MI, USA
  • Volume
    2
  • fYear
    2005
  • Firstpage
    1087
  • Abstract
    Network embedded FEC (NEF) can greatly improve the quality of service for real-time applications in p2p and proxy-based overlay networks [H. Radha and Mingquan Wu, September 2004]. Under NEF, FEC encoders and decoders (codecs) are placed in selected intermediate nodes of an overlay network The NEF codecs detect and recover lost packets, within FEC blocks at earlier stages before these blocks arrive at deeper intermediate nodes or at the final leaf nodes. This approach significantly reduces the probability of receiving undecodable FEC blocks. In many multi-hop wireless networks (such as ad hoc networks), end systems also act as routers to forward packets; higher layer functions such as FEC can then be implemented within the intermediate nodes of a multi-hop wireless network. In this paper, we analysis and compare the performance of NEF and end-to-end FEC in a multi-hop wireless channel for packet-loss recovery. We consider Markov erasure channels with memory, and hence, we present a new (and a rather simple and elegant) method for evaluating any desired loss/recovery probability measure for the Gilbert channel model. Consequently, the impact of the loss burst length on the performance of NEF and end-to-end FEC over multi-hop wireless channels are thoroughly analyzed and presented.
  • Keywords
    Markov processes; ad hoc networks; channel coding; forward error correction; statistical analysis; Gilbert channel model; Markov erasure channel; QoS; ad hoc network; loss-recovery probability; multihop wireless channel; network embedded FEC performance; p2p network; packet-loss recovery; proxy-based overlay network; quality of service; stochastic process; Electric variables measurement; Equations; Error probability; Forward error correction; Loss measurement; Markov processes; Power system modeling; Spread spectrum communication; Welding; Wireless networks;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Communications, 2005. ICC 2005. 2005 IEEE International Conference on
  • Print_ISBN
    0-7803-8938-7
  • Type

    conf

  • DOI
    10.1109/ICC.2005.1494515
  • Filename
    1494515