Page 82 - ISCI’2017
P. 82

Thus, under the implementation of a random substitution  s  we mean the specific implementation
                                                                       x
            of random vector { ( ), ( ),..., ( )}s y s y 2  s y n   and the corresponding value of the  ()s y  function we
                                   1
                                x
                                                                                               i
                                                x
                                                                                           x
                                       x
            will call the implementation of a random substitution  s  in the i-th point.
                                                                 x
                                                                                                     s y
               Independent random substitution is called such a random permutation { ( ), ( ),..., ( )},
                                                                                     s y s y
                                                                                                      x
                                                                                                2
                                                                                                         n
                                                                                         1
                                                                                      x
                                                                                            x
            for which is true
                                                 Ps y     Ps y            ( ))
                                                                           y
                                                   ( ( )) ( ( ))... ( Ps
                                         Ps x     ! n  x  1   x  2       x  n   .
                                          ( ) =
                                                    ( ( )) ( ( ))... ( ( ))
                                                ∑  Ps y  1  Ps y  2    Ps y  n
                                                               u
                                                      u
                                                                          u
                                                u= 1
                                                 1
                                                −
                  ∀
               If  iu ∈  ,  {1,2,..., }: (n  P s  u ( ))y i  =  n  then
                                             Ps y     Ps y             ( ))      n − n  1
                                                                        y
                                               ( ( )) ( ( )) ... ( Ps
                                      ( ) =
                                     Ps x     ! n  x  1   x  2        x  n   =   ! n  =                             (1)
                                                (( )) (( )) ...(( ))
                                            ∑  Ps y  1  Ps y  2     Ps y  n     ∑  n − n  ! n
                                                  u
                                                                       u
                                                           u
                                            u= 1                                u= 1
            and  s  is called the random, equiprobable and independent (or, in abbreviated form, homogeneous)
                  x
            random substitution.
               Thus, the concept of a random homogeneous substitution corresponds to a uniform probabilistic
            distribution on the set  S = { , ,..., }ss 2  s  ! n   with the independent implementation of random vectors
                                         1
                                   n
                                                                s y
                                               { ( ), ( ),..., ( )}                                                          (2)
                                                s y s y
                                                        x
                                                                    n
                                                           2
                                                                 x
                                                 x
                                                    1
                                          s ∈ S ,  x∀∈ {1,2,..., !}: ( )n  Ps = ( !)n  − 1 .
                                               n
                                           x
                                                                      x
               Modern BSC are commonly described by the random homogeneous substitution model [1, 11],
            i.e.  it  is  a  standard  assumption  that  probabilistic  properties  of  a  processed  data  blocks  bijection
            implemented by encryption function, satisfies the characteristics of a random substitution. Indeed, if
            random, equiprobable and independent selection of the master key  K () x   is associated with the choice
            of substitution  s ∈ S n ,  then the resulting ciphering  transformation will  match a random,
                              x
            equiprobable and independent comparison of ciphertext blocks to plaintext blocks on all possible
            options of bijective mapping, parameterized by key. For instance, for l-bit cipher with a  k  bit master
            key the model of random substitution will consist of subset
                                          ' S =  n  { ' , ' ,..., ' }s s  2  s  2 k  ⊂  S =  n  { , ,..., }ss 2  s  ! n
                                                 1
                                                                      1
                                                                                        l
            with random, equiprobable and  independent  2  substitutions degree  n =   2  (acting on the set
                                                            k
             Y = { ,yy 2 ,..., y 2 l }  of binary data blocks). At that  the choice  of substitution  ' s ∈  x  ' S ⊂  n  S
                   1
                                                                                                            n
            (inplementation of random vector { ( ), ( ),..., (s y s y 2  s y 2 l )}) is set randomly, with equal probability
                                                  1
                                               x
                                                               x
                                                      x
            and independently of selected  k -bit master key  K ()x   value.
            82
   77   78   79   80   81   82   83   84   85   86   87