Abstract:The classical Brun--Titchmarsh theorem gives an upper bound, which is of correct order of magnitude in the full range, for the number of primes $p\leqslant x$ satisfying $p\equiv a\bmod q$. We strength this inequality for $\log q/\log x$ in different ranges, improving upon previous works by Motohashi, Goldfeld, Iwaniec, Friedlander and Iwaniec, and Maynard for general or special moduli. In particular, we obtain a Burgess-like constant in the full range $q<x^{1/2-}.$ The proof is based on various estimates for character and exponential sums, by appealing to arithmetic exponent pairs and bilinear forms with algebraic trace functions from $\ell$-adic cohomology, and sums of Kloosterman sums from spectral theory of automorphic forms, as well as large value theorem for Dirichlet polynomials due to Huxley.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is to improve the upper bound estimate in the classical Brun - Titchmarsh theorem. Specifically, the Brun - Titchmarsh theorem gives an upper bound estimate for the number of primes \(p\leq x\) satisfying \(p\equiv a\pmod{q}\):
\[
\pi(x;q, a):=\left|\{p\leq x:p\equiv a\pmod{q}\}\right|
\]
The classical Brun - Titchmarsh theorem shows that:
\[
\max_{(a,q) = 1}\pi(x;q, a)\ll\frac{x}{\varphi(q)\log(x/q)}
\]
where \(q < x\), and \(\varphi(q)\) is the Euler function.
The main purpose of this paper is to optimize the constant \(C(\varpi)\) in this inequality in different ranges, such that:
\[
\max_{(a,q) = 1}\pi(x;q, a)\leq\left(C(\varpi)+\varepsilon\right)\frac{x}{\varphi(q)\log x}
\]
holds for any \(\varepsilon>0\) and \(x > x_0(\varepsilon)\). By introducing new methods and techniques, especially by using various estimates of character sums and exponential sums, arithmetic exponent pairs, trace functions and Kloosterman sums and other tools, the author improves the previous best results in different ranges.
For example, for \(\varpi\in\left(\frac{9}{20},\frac{1}{2}\right)\), the author obtains a new Burgess - type constant:
\[
C(\varpi)=\frac{16}{8-(3 + 2\vartheta)\varpi}
\]
where \(\vartheta\) is the exponent of the Ramanujan - Petersson conjecture. In addition, for a wider range of moduli \(q\), the author also provides new options, especially in the case where \(q\) is a prime, and these results are independent of the Ramanujan - Petersson conjecture.
In summary, the main contribution of this paper is to further optimize the upper bound estimate in the Brun - Titchmarsh theorem by introducing new mathematical tools and techniques, thereby providing more accurate results in the field of number theory.