• DocumentCode
    896919
  • Title

    Six-dimensional trellis-coding with QAM signal sets

  • Author

    Chouly, Antoine ; Sari, Hikmet

  • Author_Institution
    Lab. d´´Electrotech. Philips, Limeil-Brevannes, France
  • Volume
    40
  • Issue
    1
  • fYear
    1992
  • fDate
    1/1/1992 12:00:00 AM
  • Firstpage
    24
  • Lastpage
    33
  • Abstract
    A family of 6-D trellis-coded modulation (TCM) schemes which involve a 2-step partitioning of the constituent QAM signal alphabet is presented. With infinite constellations without shaping, the asymptotic coding gain is 3 dB for the 2-state code, 4 dB for the 4- and 8-state codes, and 5 dB for the 16- and 32-state codes which involve a smaller alphabet expansion. The authors also describe a rotationally invariant 16-state code that achieves the same asymptotic gain as its linear counterpart. Practical signal constellations are described for 6-D TCM with the spectral efficiency of uncoded 64-QAM, and the performance of these schemes is studied by means of computer simulations. It was found that they achieve an additional coding gain of 0.2-0.3 dB over infinite hypercube-type constellations. The performance of the presented schemes at practical signal-to-noise ratio values is evaluated using transfer function techniques
  • Keywords
    amplitude modulation; codes; encoding; 16-state code; 2-state code; 32-state codes; 4-state code; 6-D trellis-coded modulation; 8-state codes; QAM signal sets; SNR; TCM; asymptotic coding gain; computer simulations; rotationally invariant code; signal constellations; signal-to-noise ratio; spectral efficiency; transfer function; trellis coding; Constellation diagram; Convolutional codes; Gain; Linear code; Multidimensional systems; Phase shift keying; Quadrature amplitude modulation; Signal to noise ratio; Transfer functions; Two dimensional displays;
  • fLanguage
    English
  • Journal_Title
    Communications, IEEE Transactions on
  • Publisher
    ieee
  • ISSN
    0090-6778
  • Type

    jour

  • DOI
    10.1109/26.126704
  • Filename
    126704