• DocumentCode
    1265941
  • Title

    The decimation-Hadamard transform of two-level autocorrelation sequences

  • Author

    Gong, Guang ; Golomb, Solomon W.

  • Author_Institution
    Dept. of Electr. & Comput. Eng., Waterloo Univ., Ont., Canada
  • Volume
    48
  • Issue
    4
  • fYear
    2002
  • fDate
    4/1/2002 12:00:00 AM
  • Firstpage
    853
  • Lastpage
    865
  • Abstract
    A new method to study and search for two-level autocorrelation sequences for both binary and nonbinary cases is developed. This method iteratively applies two operations: decimation and the Hadamard transform based on general orthogonal functions, referred to as the decimation-Hadamard transform (DHT). The second iterative DHT can transform one class of such sequences into another inequivalent class of such sequences, a process called realization. The existence and counting problems of the second iterative DHT are discussed. Using the second iterative DHT, and starting with a single binary m-sequence (when n is odd), we believe one can obtain all the known two-level autocorrelation sequences of period 2n-1 which have no subfield factorization. We have verified this for odd n⩽17. Interestingly, no previously unknown examples were found by this process for any odd n⩽17. This is supporting evidence (albeit weak) for the conjecture that all families of cyclic Hadamard difference sets of period 2n -1 having no subfield factorization are now known, at least for odd n. Experimental results are provided
  • Keywords
    Hadamard transforms; correlation theory; iterative methods; m-sequences; DHT; binary m-sequence; cyclic Hadamard difference sets; decimation-Hadamard transform; iteration; orthogonal functions; realization; two-level autocorrelation sequences; Autocorrelation; Binary sequences; Cryptography; Galois fields; Helium;
  • fLanguage
    English
  • Journal_Title
    Information Theory, IEEE Transactions on
  • Publisher
    ieee
  • ISSN
    0018-9448
  • Type

    jour

  • DOI
    10.1109/18.992772
  • Filename
    992772