• DocumentCode
    3272343
  • Title

    Distributed source coding with cyclic codes and their duals

  • Author

    Cao, Xiaomin ; Kuijper, Margreta

  • Author_Institution
    Dept. of Electr. & Electron. Eng., Univ. of Melbourne, Melbourne, VIC, Australia
  • fYear
    2009
  • fDate
    8-10 Dec. 2009
  • Firstpage
    1
  • Lastpage
    5
  • Abstract
    We consider the problem of compression of multiple correlated sources. It is assumed that the sources are given as a sequence with neighbouring sources correlated in terms of Hamming distance. In previous work we derived a general distributed source coding (DSC) framework that uses the parity check matrix of a suitably chosen block code C1. In this paper we focus on the special case that C1 is a cyclic code. This allows us to formulate our DSC framework in terms of polynomials. The polynomial structure has the advantage that it allows for implementation via efficient shift registers. We show that the Sylvester resultant of polynomials plays a crucial role in the DSC framework. We address the notion of ¿complementarity¿ of codes and specify the DSC framework in terms of the cyclic code C1 and the reciprocal C2 of its dual code for the special case that C2 is complementary to C1. We illustrate our polynomial DSC framework for a sequence of 3 sources through a (15, 7) BCH code C1 and the reciprocal of its dual code C2.
  • Keywords
    block codes; cyclic codes; matrix algebra; parity check codes; source coding; Hamming distance; block code; cyclic codes; distributed source coding; multiple correlated sources; parity check matrix; Block codes; Decoding; Hamming distance; Image coding; Image sensors; Parity check codes; Polynomials; Sensor phenomena and characterization; Shift registers; Source coding;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Information, Communications and Signal Processing, 2009. ICICS 2009. 7th International Conference on
  • Conference_Location
    Macau
  • Print_ISBN
    978-1-4244-4656-8
  • Electronic_ISBN
    978-1-4244-4657-5
  • Type

    conf

  • DOI
    10.1109/ICICS.2009.5397707
  • Filename
    5397707