• DocumentCode
    3172941
  • Title

    On lexicographic max-min node lifetime for wireless sensor networks

  • Author

    Hou, Y. Thomas ; Shi, Yi ; Sherali, Hanif D.

  • Author_Institution
    Bradley Dept. of Electr. & Comput. Eng., Virginia Tech., Blacksburg, VA, USA
  • Volume
    7
  • fYear
    2004
  • fDate
    20-24 June 2004
  • Firstpage
    3790
  • Abstract
    We study the network lifetime problem by considering not only the maximized time until the first node fails, but also the maximized lifetime for all the nodes in the network which we define as the lexicographic max-min (LMM) node lifetime problem. The main contributions of this paper are two-fold. First, we develop a polynomial-time algorithm to derive the LMM-optimal node lifetime vector, which effectively circumvents the computational complexity problem associated with an existing state-of-the-art approach, which is exponential. Second, we present a simple (also polynomial-time) algorithm to calculate the flow routing schedule such that the LMM-optimal node lifetime vector can be achieved. Our results in this paper advance the state-of-the-art algorithmic design to network-wide node lifetime problems.
  • Keywords
    computational complexity; minimax techniques; telecommunication network routing; wireless sensor networks; computational complexity; energy constraint; flow routing schedule; lexicographic max-min node lifetime; node lifetime vector; polynomial-time algorithm; power control; state-of-the-art algorithmic design; wireless sensor networks; Computational complexity; Computer industry; Computer networks; Polynomials; Routing; Scheduling algorithm; Surveillance; Systems engineering and theory; Time measurement; Wireless sensor networks;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Communications, 2004 IEEE International Conference on
  • Print_ISBN
    0-7803-8533-0
  • Type

    conf

  • DOI
    10.1109/ICC.2004.1313262
  • Filename
    1313262