• DocumentCode
    1355142
  • Title

    Power allocation for coded OFDM via linear programming

  • Author

    Kenarsari-Anhari, Alireza ; Lampe, Lutz

  • Author_Institution
    Dept. of Electr. & Comput. Eng., Univ. of British Columbia, Vancouver, BC, Canada
  • Volume
    13
  • Issue
    12
  • fYear
    2009
  • fDate
    12/1/2009 12:00:00 AM
  • Firstpage
    887
  • Lastpage
    889
  • Abstract
    The combination of bit-interleaved coded modulation and orthogonal frequency-division multiplexing (BICOFDM) forms a powerful coded modulation scheme for transmission over wideband channels. Recently, Moon and Cox presented a new power allocation method to minimize the biterror rate (BER) of BIC-OFDM. It requires the solution of a convex optimization problem and is limited to (complex) binary transmission. Motivated by their work, in this letter we present an alternative power allocation method, which has the advantages of being a linear program and applicable to arbitrary signal constellations. Our approach relies on a BER approximation which becomes tight for asymptotically large signal-to-noise ratios. Simulative evidence shows that the proposed power allocation method achieves a performance very close to that from for the case of quadrature phase-shift keying.
  • Keywords
    OFDM modulation; convex programming; error statistics; interleaved codes; linear programming; quadrature phase shift keying; BER approximation; bit-interleaved coded modulation; coded OFDM; convex optimization problem; linear programming; orthogonal frequency-division multiplexing; power allocation; quadrature phase-shift keying; Bit error rate; Constellation diagram; Frequency division multiplexing; Frequency modulation; Interleaved codes; Linear programming; Modulation coding; Moon; OFDM modulation; Wideband; Power loading, bit-interleaved coded modulation (BICM), orthogonal frequency division multiplexing (OFDM), linear program;
  • fLanguage
    English
  • Journal_Title
    Communications Letters, IEEE
  • Publisher
    ieee
  • ISSN
    1089-7798
  • Type

    jour

  • DOI
    10.1109/LCOMM.2009.12.091590
  • Filename
    5353252