A simpler proof of Levinson's theorem (Q3216737)
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 3882533
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | A simpler proof of Levinson's theorem |
scientific article; zbMATH DE number 3882533 |
Statements
A simpler proof of Levinson's theorem (English)
0 references
1985
0 references
mean-value theorem
0 references
Levinson's theorem
0 references
Riemann zeta-function
0 references
zeros
0 references
critical line
0 references
Dirichlet polynomial
0 references
large sieve
0 references
0 references
0 references
0 references
0.7873972
0 references
0.78059757
0 references
0.7743163
0 references
\textit{N. Levinson} [Adv. Math. 13, 383--436 (1974; Zbl 0281.10017)] proved that at least one third of the non-trivial zeros of \(\zeta(s)\) lie on the critical line. The proof was decidedly long -- 53 printed pages. The key quantity to be calculated is a mean value \[ I=\int^{T+U}_{T}| \psi G(a+it)|^ 2 \,dt, \] where \(U=T(\log T)^{-10}\), \(G(s)=\zeta '(1-s)\) and \(\psi(s)\) is a suitable Dirichlet polynomial \(\sum b(k)k^{-s}\) of length \(T^{1/2-\varepsilon}\). In Levinson's treatment \(G\) is expressed via the approximate functional equation, so that \(I\) is the sum of three integrals involving Dirichlet polynomials. The most troublesome of these also involves a factor \(\chi(s)\). The three integrals are estimated separately, and when the results are added a considerable cancellation occurs. The purpose of this paper is to give an alternative method for evaluating \(I\) which, though by no means easy, is distinctly shorter. It also has the advantage of producing the leading terms without the surprising cancellations of Levinson's method. The new approach is more analytical and uses the large sieve.
0 references