English

On the weight distributions of several classes of cyclic codes from APN monomials

Information Theory 2013-08-28 v1 math.IT

Abstract

Let m3m\geq 3 be an odd integer and pp be an odd prime. % with p1=2rhp-1=2^rh, where hh is an odd integer. In this paper, many classes of three-weight cyclic codes over Fp\mathbb{F}_{p} are presented via an examination of the condition for the cyclic codes C(1,d)\mathcal{C}_{(1,d)} and C(1,e)\mathcal{C}_{(1,e)}, which have parity-check polynomials m1(x)md(x)m_1(x)m_d(x) and m1(x)me(x)m_1(x)m_e(x) respectively, to have the same weight distribution, where mi(x)m_i(x) is the minimal polynomial of πi\pi^{-i} over Fp\mathbb{F}_{p} for a primitive element π\pi of Fpm\mathbb{F}_{p^m}. %For p=3p=3, the duals of five classes of the proposed cyclic codes are optimal in the sense that they meet certain bounds on linear codes. Furthermore, for p3(mod4)p\equiv 3 \pmod{4} and positive integers ee such that there exist integers kk with gcd(m,k)=1\gcd(m,k)=1 and τ{0,1,,m1}\tau\in\{0,1,\cdots, m-1\} satisfying (pk+1)e2pτ(modpm1)(p^k+1)\cdot e\equiv 2 p^{\tau}\pmod{p^m-1}, the value distributions of the two exponential sums T(a,b)=xFpmω\Tr(ax+bxe)T(a,b)=\sum\limits_{x\in \mathbb{F}_{p^m}}\omega^{\Tr(ax+bx^e)} and S(a,b,c)=xFpmω\Tr(ax+bxe+cxs), S(a,b,c)=\sum\limits_{x\in \mathbb{F}_{p^m}}\omega^{\Tr(ax+bx^e+cx^s)}, where s=(pm1)/2s=(p^m-1)/2, are settled. As an application, the value distribution of S(a,b,c)S(a,b,c) is utilized to investigate the weight distribution of the cyclic codes C(1,e,s)\mathcal{C}_{(1,e,s)} with parity-check polynomial m1(x)me(x)ms(x)m_1(x)m_e(x)m_s(x). In the case of p=3p=3 and even ee satisfying the above condition, the duals of the cyclic codes C(1,e,s)\mathcal{C}_{(1,e,s)} have the optimal minimum distance.

Keywords

Cite

@article{arxiv.1308.5885,
  title  = {On the weight distributions of several classes of cyclic codes from APN monomials},
  author = {Chunlei Li and Nian Li and Tor Helleseth and Cunsheng Ding},
  journal= {arXiv preprint arXiv:1308.5885},
  year   = {2013}
}