Self-orthogonal codes from p-divisible codes

The self-orthogonality and divisibility are two important properties of linear codes. It is interesting to establish relationship between them. By the well-known Gleason-Pierce-Ward Theorem, all self-dual divisible codes have been totally classified. However, the relationship between the self-orthogonality and divisibility of a
q
-ary linear codes is known only for
q
= 2, 3 by Huffman and Pless in 2003. It has remained open for more than 20 years to consider other cases. The purpose of this paper is to settle this open problem under certain conditions and construct new families of self-orthogonal codes. Let
q
be a power of an odd prime
p
. Firstly, we prove that any
p
-divisible code containing the all-1 vector over the finite field F
_q
is self-orthogonal. More generally, it is concluded that any
p
-divisible [
n
,
k
] linear code over F
_q
containing codewords of weight
n
is monomially equivalent to an [
n
,
k
] self-orthogonal code over F
_q
. This result provides a very efficient way to find self-orthogonal codes from
p
-divisible codes. Secondly, we apply this result to construct self-orthogonal codes with excellent parameters or nice applications. For one thing, we use this result to study the self-orthogonality of generalized Reed-Muller codes, certain projective two-weight codes, and Griesmer codes. For another thing, by this useful result as well as the extending and augmentation techniques for linear codes, we construct eight new families of self-orthogonal divisible codes. These self-orthogonal codes and their duals contain many optimal or almost optimal codes. Besides, some self-orthogonal codes support combinatorial designs and some of them are proved to be optimal or almost optimal locally recoverable codes.