• DocumentCode
    2841524
  • Title

    Algorithms for iterative phase noise estimation based on a truncated DCT expansion

  • Author

    Bhatti, Jabran ; Noels, Nele ; Moeneclaey, Marc

  • Author_Institution
    Dept. of Telecommun. & Inf. Process., Ghent Univ., Ghent, Belgium
  • fYear
    2011
  • fDate
    26-29 June 2011
  • Firstpage
    51
  • Lastpage
    55
  • Abstract
    We present two phase noise estimation algorithms for single-carrier burst communications. The first and second estimation techniques (denoted as PEA1 and PEA2 respectively) are based on a truncated discrete cosine transform (DCT) basis expansion of the phase noise and of the corresponding phasor, respectively and do not require any knowledge of the phase noise statistics. An initial pilot-based estimate is iteratively improved by means of the expectation-maximization algorithm, yielding a soft-decision-directed phase noise estimate. The performances of PEA1 and PEA2 are compared in terms of the mean-square phase error and bit-error rate. Numerical results show that PEA2 outperforms PEA1 for practical values of the signal-to-noise ratio.
  • Keywords
    channel estimation; discrete cosine transforms; error statistics; expectation-maximisation algorithm; phase noise; telecommunication networks; bit-error rate; expectation-maximization algorithm; iterative phase noise estimation algorithm; signal-to-noise ratio; single-carrier burst communications; truncated DCT expansion; truncated discrete cosine transform; Algorithm design and analysis; Bit error rate; Discrete cosine transforms; Estimation; Phase estimation; Phase noise; Signal processing algorithms;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Signal Processing Advances in Wireless Communications (SPAWC), 2011 IEEE 12th International Workshop on
  • Conference_Location
    San Francisco, CA
  • ISSN
    1948-3244
  • Print_ISBN
    978-1-4244-9333-3
  • Electronic_ISBN
    1948-3244
  • Type

    conf

  • DOI
    10.1109/SPAWC.2011.5990465
  • Filename
    5990465