In applied mathematics, the Johnson bound (named after Selmer Martin Johnson) is a limit on the size of error-correcting codes, as used in coding theory for data transmission or communications.
Definition
Let  be a q-ary code of length
 be a q-ary code of length  , i.e. a subset of
, i.e. a subset of  . Let
. Let  be the minimum distance of
 be the minimum distance of  , i.e.
, i.e.
 
where  is the Hamming distance between
 is the Hamming distance between  and
 and  .
.
Let  be the set of all q-ary codes with length
 be the set of all q-ary codes with length  and minimum distance
 and minimum distance  and let
 and let  denote the set of codes in
 denote the set of codes in  such that every element has exactly
 such that every element has exactly  nonzero entries.
 nonzero entries.
Denote by  the number of elements in
 the number of elements in  . Then, we define
. Then, we define  to be the largest size of a code with length
 to be the largest size of a code with length  and minimum distance
 and minimum distance  :
:
 
Similarly, we define  to be the largest size of a code in
 to be the largest size of a code in  :
:
 
Theorem 1 (Johnson bound for  ):
):
If  ,
,
 
If  ,
,
 
 Theorem 2 (Johnson bound for  ):
):
(i) If  
 
(ii) If  , then define the variable
, then define the variable  as follows. If
 as follows. If  is even, then define
 is even, then define  through the relation
 through the relation  ; if
; if  is odd, define
 is odd, define  through the relation
 through the relation  .  Let
.  Let  . Then,
. Then,
 
where  is the floor function.
 is the floor function.
Remark: Plugging the bound of Theorem 2 into the bound of Theorem 1 produces a numerical upper bound on  .
.
See also
References