• DocumentCode
    3267184
  • Title

    An Optimum pre-filter for ICA based mulit-input multi-output OFDM System

  • Author

    Khosravy, Mahdi ; Alsharif, Mohammad Reza ; Khosravi, Mohammad ; Yamashita, Katsumi

  • Author_Institution
    Inf. Dept., Univ. of the Ryukyus, Nishihara, Japan
  • Volume
    5
  • fYear
    2010
  • fDate
    22-24 June 2010
  • Abstract
    This paper presents an Optimum pre-filter for ICA based mulit-input multi-output OFDM System. In our former work, we presented a pre-filter solution for multiuser reconstruction in ICA based MIMO-OFDM systems. Here, based on the structure of the system and former pre-filter solution, an optimization problem has been designed to give us the optimum pre-filter. The designed problem is based on minimization of the estimation error of ambiguities inherent to ICA in reconstruction of the data after separation. The optimization problem is mathematically solved and a set of optimized coefficients are obtained. Among the obtained set as optimum coefficients, optimum coefficients are acquired in a way not only minimize the estimation error of multiuser data reconstruction, but also lead the least error probability in detection of demodulated symbols.
  • Keywords
    MIMO communication; OFDM modulation; error statistics; filtering theory; independent component analysis; signal reconstruction; ICA-based MIMO-OFDM System; demodulated symbol detection; error probability; estimation error minimisation; independent component analysis; multiple-input multiple-output system; multiuser data reconstruction; optimization problem; optimum coefficients; optimum prefilter; Blind equalizers; Design optimization; Educational technology; Error probability; Estimation error; Finite impulse response filter; Independent component analysis; MIMO; OFDM; Source separation; Blind equalization; Independent component analysis; MIMO systems; OFDM;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Education Technology and Computer (ICETC), 2010 2nd International Conference on
  • Conference_Location
    Shanghai
  • Print_ISBN
    978-1-4244-6367-1
  • Type

    conf

  • DOI
    10.1109/ICETC.2010.5529798
  • Filename
    5529798