• DocumentCode
    3862461
  • Title

    Complex-Valued Maximum Joint Entropy Algorithm for Blind Decision Feedback Equalizer

  • Author

    Vladimir R. Krstic;Zoran Petrovic

  • Author_Institution
    Institute Mihailo Pupin, Volgina 15, 11060 Belgrade, Serbia, E-mail: vladak@kondor.imp.bg.ac.yu
  • fYear
    2007
  • Firstpage
    601
  • Lastpage
    604
  • Abstract
    This paper proposes the complex domain solution of the blind decision feedback equalizer (DFE) based on joint entropy maximization (JEM) criterion. For the feedback filter of DFE, a new complex activation function is defined and then the stochastic gradient algorithm of the JEM type is derived. The fact that the JEM cost is a real function of complex quantities suggests the method associated to a least man-square criterion.
  • Keywords
    "Entropy","Decision feedback equalizers","Cost function","Stochastic processes","Nonlinear filters","Neural networks","Multi-layer neural network","Multilayer perceptrons","Blind equalizers","Neurofeedback"
  • Publisher
    ieee
  • Conference_Titel
    Telecommunications in Modern Satellite, Cable and Broadcasting Services, 2007. TELSIKS 2007. 8th International Conference on
  • Print_ISBN
    978-1-4244-1467-3
  • Type

    conf

  • DOI
    10.1109/TELSKS.2007.4376086
  • Filename
    4376086
    • Link To Document
    https://search.isc.ac/dl/search/defaultta.aspx?DTC=49&DC=3862461