English

On solving a restricted linear congruence using generalized Ramanujan sums

Number Theory 2017-08-16 v1

Abstract

Consider the linear congruence equation x1++xkb(mod n)x_1+\ldots+x_k \equiv b\,(\text{mod } n) for b,nZb,n\in\mathbb{Z}. By (a,b)s(a,b)_s, we mean the largest lsNl^s\in\mathbb{N} which divides aa and bb simultaneously. For each djnd_j|n, define Cj,s={1xns(x,ns)s=djs}\mathcal{C}_{j,s} = \{1\leq x\leq n^s | (x,n^s)_s = d^s_j\}. Bibak et al. gave a formula using Ramanujan sums for the number of solutions of the above congruence equation with some gcd restrictions on xix_i. We generalize their result with generalized gcd restrictions on xix_i by proving that for the above linear congruence, the number of solutions is 1nsdncd,s(b)j=1τ(n)(cndj,s(nsds))gj\frac{1}{n^s}\sum\limits_{d|n}c_{d,s}(b)\prod\limits_{j=1}^{\tau(n)}\left(c_{\frac{n}{d_j},s}(\frac{n^s}{d^s})\right)^{g_j} where gj={x1,,xk}Cj,sg_j = |\{x_1,\ldots, x_k\}\cap \mathcal{C}_{j,s}| for j=1,τ(n)j=1,\ldots \tau(n) and cd,sc_{d,s} denote the generalized ramanujan sum defined by E. Cohen.

Keywords

Cite

@article{arxiv.1708.04505,
  title  = {On solving a restricted linear congruence using generalized Ramanujan sums},
  author = {K Vishnu Namboothiri},
  journal= {arXiv preprint arXiv:1708.04505},
  year   = {2017}
}
R2 v1 2026-06-22T21:15:07.369Z