• DocumentCode
    1035678
  • Title

    Iterative detection/decoding for two-track partial response channels

  • Author

    Ma, Xiao ; Ping, Li

  • Author_Institution
    Dept. of Electron. Eng., City Univ. of Hong Kong, Kowloon, China
  • Volume
    8
  • Issue
    7
  • fYear
    2004
  • fDate
    7/1/2004 12:00:00 AM
  • Firstpage
    464
  • Lastpage
    466
  • Abstract
    In this letter, we investigate the iterative detection/decoding algorithms for two-track partial response channels. We first describe the iterative algorithm as an iterative message-passing/processing algorithm over a high-level normal graph and then propose two trellis-based detection algorithms. The natural one is based on a joint trellis, which can be implemented for low-order channels. For moderate-order channels, we propose a separated detection, which is based on two single-track trellises. Compared with the joint detection, the separated detection is simpler in most cases. Simulation results show that, for two-track EPR4 channels with minor intertrack interferences, the separated detection causes only marginal degradation in performance.
  • Keywords
    iterative decoding; message passing; trellis codes; turbo codes; concatenated system; high-level normal graph; intertrack interference; iterative detection-decoding; iterative message-passing-processing algorithm; multitrack recording; partial response signaling; trellis-based detection algorithms; turbo equalization; two-track partial response channels; Covariance matrix; Degradation; Detection algorithms; Error correction codes; Intersymbol interference; Iterative algorithms; Iterative decoding; Maximum likelihood decoding; Maximum likelihood detection; Partial response channels; Concatenated system; ITI; PR; intertrack interference; multitrack recording; partial response; signaling; turbo equalization;
  • fLanguage
    English
  • Journal_Title
    Communications Letters, IEEE
  • Publisher
    ieee
  • ISSN
    1089-7798
  • Type

    jour

  • DOI
    10.1109/LCOMM.2004.832749
  • Filename
    1315678