A new mean-value result for Dirichlet L-functions and polynomials (Q4652173)
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 2138384
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | A new mean-value result for Dirichlet L-functions and polynomials |
scientific article; zbMATH DE number 2138384 |
Statements
24 February 2005
0 references
Dirichlet \(L\)-functions
0 references
Dirichlet polynomials
0 references
mean value
0 references
sums of Kloosterman sums
0 references
A new mean-value result for Dirichlet L-functions and polynomials (English)
0 references
0.9450445
0 references
0.9310583
0 references
0.9310583
0 references
0.92256224
0 references
0.92256224
0 references
0.9155585
0 references
0.91264665
0 references
0 references
0.9025652
0 references
0 references
In many problems of multiplicative number theory bounds for mean values of Dirichlet polynomials are of paramount importance. In this paper \textit{N. Watt}'s [J. Number Theory 53, No. 1, 179--210 (1995; Zbl 0837.11050)] result \[ \int_{-T}^{T}| \zeta(\tfrac12+it)| ^4 \Bigl| \sum_{m\leq M}a_{m}m^{-it}\Bigr|^2 \, dt \ll MT^{1+\varepsilon}(1+M^{2}T^{-{1\over 2}})\,\max_{m\leq M} | a_{m}| ^2 \] is generalized to \[ \begin{multlined} {1\over \phi(q)}\sum_{\chi\pmod q} \int_{-T}^{T}| L(\tfrac12+it,\chi)| ^4 \Bigl| \sum_{m\leq M}a_{m}\chi(m)m^{-it}\Bigr| ^2 \, dt\\ \ll (MT)^{1+\varepsilon}(1+M^{2}T^{-{1\over 2}}q^{-\gamma})\,\max_{m\leq M} | a_{m}| ^2 , \end{multlined} \] where \(M\geq 1,\, T\geq q^{3/5}\) and \(\gamma = {1\over 2}-{7\over 64}\). The methods of this paper also give \[ \begin{multlined} \sum_{q\leq Q} {1\over \phi(q)}\sum_{\chi(\bmod \, q)} \int_{-T}^{T}| L(\tfrac12+it,\chi)| ^4 \Bigl| \sum_{m\leq M}a_{m}\chi(m)m^{-it}\Bigr|^2 \, dt\\ \ll (MTQ)^{1+\varepsilon}(1+M^{2}(TQ)^{-{1\over 2}})\,\max_{m\leq M}| a_{m}| ^2 \end{multlined} \] for \(M\geq 1\) and \(T\geq Q^{3/5}\). It is remarked that conditions of the kind \(T\geq q^{3/5}\) can be somewhat relaxed at the expense of having more complicated expressions for the upper bound, but some condition \(T\geq q^{\delta},\, \delta >0\), is necessary to make the proofs work. As an application the authors have the following result: There is an absolute constant \(\delta >0\) such that, for all large \(x\) and for almost all \(q\leq x^{\delta}\) (the number of exceptional \(q\) being \(O(x^{\delta}(\log x)^{-A})\) for any \(A>0\)), every arithmetic progression \(a \, \bmod \, q\) with \((a,q)=1\) contains primes from the interval \((x-x^{.525},x]\). This application uses \textit{A. Kumchev}'s method and improves his result in [Q. J. Math. 53, No. 4, 479--501 (2002; Zbl 1045.11061)]. The proof begins by some simplifications and appeals to several results from \textit{H. L. Montgomery}'s book [Topics in multiplicative number theory, Lecture Notes in Mathematics 227. Berlin: Springer Verlag (1971; Zbl 0216.03501)] including A. F. Lavrik's version of replacing \(| L(s,\chi)| ^4\) with a Dirichlet polynomial. The track laid out in Watt's argument is followed until a generalization of a proposition on a weighted average of Kloosterman sums is needed. By the Kuznetsov trace formula, sums of Kloosterman sums are related to sums of Fourier coefficients of Eisenstein series and cusp-forms. An extra summation condition occurring in this generalization brings the main challenge faced in this paper. The new generalized proposition includes an improvement on the old one due to the use of work of \textit{H. H. Kim} (and its appendix co-authored by \textit{P. Sarnak}) [J. Am. Math. Soc. 16, No. 1, 139--183 (2003; Zbl 1018.11024)] and reflects the progress towards Selberg's eigenvalue conjecture achieved there.
0 references