• DocumentCode
    1379639
  • Title

    The hardness of solving subset sum with preprocessing

  • Author

    Lobstein, Antoine

  • Author_Institution
    CNRS, Paris, France
  • Volume
    36
  • Issue
    4
  • fYear
    1990
  • fDate
    7/1/1990 12:00:00 AM
  • Firstpage
    943
  • Lastpage
    946
  • Abstract
    The two problems subset sum and linear decoding (which have given birth to the knapsack and the McEliece public-key cryptosystems) are known to be NP-complete. For linear decoding, it has been proved that, even if one knows the linear code in advance and can preprocess it, the existence of a polynomial-time decoding algorithm would imply that the polynomial-time hierarchy collapses at an early stage. It is proved that the same holds for subset sum. Even if the knapsack is known in advance and can be preprocessed, there is no polynomial-time algorithm solving it, unless the polynomial hierarchy collapses. Also given is the sketch of a new, straightforward proof of this result for linear decoding
  • Keywords
    cryptography; decoding; McEliece public-key cryptosystems; NP-complete; knapsack cryptosystem; linear decoding; polynomial hierarchy; polynomial-time decoding algorithm; preprocessing; subset sum; Block codes; Decoding; Linear code; NP-complete problem; Parity check codes; Polynomials; Public key; Public key cryptography; Symmetric matrices; Vectors;
  • fLanguage
    English
  • Journal_Title
    Information Theory, IEEE Transactions on
  • Publisher
    ieee
  • ISSN
    0018-9448
  • Type

    jour

  • DOI
    10.1109/18.53763
  • Filename
    53763