Generalizations of a Classical Theorem in Number Theory (Q4116501)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

scientific article; zbMATH DE number 3540867
Language Label Description Also known as
default for all languages
No label defined
    English
    Generalizations of a Classical Theorem in Number Theory
    scientific article; zbMATH DE number 3540867

      Statements

      0 references
      1976
      0 references
      Generalizations of a Classical Theorem in Number Theory (English)
      0 references
      From the class number formula of Dirichlet follows \[ \sum_{k=1}^{\frac{p-1}{2}} \left(\frac{k}{p}\right)>0, \] where \(p\) is a prime with \(4\mid p+1\) and \(\left(\frac{k}{p}\right)\) denotes the Legendre symbol. Therefore, if \(p\) is an odd prime, \(p\equiv 3\pmod 4\), then there exist more quadratic residues mod \(p\) between \(0\) and \(p/2\) than between \(p/2\) and \(p\). (A simple analytic proof of this theorem is given by \textit{L. Moser} [Proc. Am. Math. Soc. 2, 503--504 (1951; Zbl 0045.02001)].) This is equivalent to the assertion that for an odd prime \(p\), the sum of the quadratic residues in \((0,p)\) is strictly less than the sum of the quadratic nonresidues in \((0,p)\) if and only if \(p\equiv 3\pmod 4\). In this paper generalizations of this theorem are considered. Firstly, the problem of \(k\)-th power residues and nonresidues (mod \(p\)), \(k>2\), is considered. Secondly, a generalization to quadratic residues and nonresidues (mod \(n\)), \(n\) an arbitrary integer \(>2\), is given. From the set of least positive residues (mod \(n\)), let \(c_0\) denote the subgroup of quadratic residues (mod \(n\)) and let \(c_1,c_2,\dots,c_t\) be the cosets which can be formed with respect to this subgroup. The conjecture, that the sum of the elements,in \(c_0\) is less than or equal to the sum in any of the other cosets for every integer \(n>2\), is in general false if we have \(k>2\). Counterexamples are given \((p=29\), \(k=4\) etc.). For \(k=2\) the following theorems are proved. (1) If \(n=p^\alpha\), \(p\) odd, then the sum of the quadratic nonresidues (mod \(n\)) exceeds the sum of quadratic residues (mod \(n\)) if and only if \(p\equiv 3\pmod 4\). (2) If \(n =2^\alpha\), \(\alpha\geq 1\), no two coset sums have the same value. Moreover, if \(\alpha\geq 3\) the coset sums can be arranged in strictly increasing order with common difference \(2^{\alpha-2}\). The smallest of these four coset sums is the sum in \(c_0\) which is precisely \(2^{\alpha-3}(2^{\alpha-1}-3)\) for all \(\alpha\geq 3\). (3) If \(n = 2^{\alpha_0}p_1^{\alpha_1}p_2^{\alpha_2}\cdots p_r^{\alpha_r}\) is the prime factorization of \(n\), then there are \(2^{r+\beta-1}\) cosets which can be formed with respect to the subgroup of quadratic residues (mod \(n\)); \(\beta =1\) if \(\alpha_0 =0\) or \(1\), \(\beta =2\) if \(\alpha_0 =2\), \(\beta =3\) if \(\alpha_0\geq 3\). If the coset sums are all equal, each must have the value \((n\cdot \varphi(n)) / 2^{r+\beta}\). (4) If \(n = 2^{\alpha_0}p_1^{\alpha_1}p_2^{\alpha_2}\cdots p_r^{\alpha_r}\) with \(p_i \equiv 1\pmod 4\) for all \(i\), \(1\leq i\leq r\), and if \(\alpha_0 =0\) or \(1\), then the sum in each coset is \((n\cdot \varphi(n)) / 2^{r+1}\). (5) If \(n = 2^{\alpha_0}p_1^{\alpha_1}p_2^{\alpha_2}\cdots p_r^{\alpha_r}\) with \(p_i \equiv 3\pmod 4\) for all \(i\), \(1\leq i\leq r\), then the coset sums are not identical. (6) If \(n= 2p^\alpha\), \(\alpha\geq 1\), \(p\equiv 3\pmod 8\), then the sum of the quadratic nonresidues (mod \(n\)) exceeds the sum of quadratic residues (mod \(n\)). (7) If \(n= 2p^\alpha\), \(\alpha\geq 1\), \(p\) an odd prime \(\not\equiv 3\pmod 8\), then the sum of the quadratic residues (mod \(n\)) and the sum of the quadratic nonresidues (mod \(n\)) is exactly \((n\cdot \varphi(n))/4\).
      0 references

      Identifiers