On sums of squares of primes and a \(k\)-th power of prime (Q1712577)

From MaRDI portal
scientific article
Language Label Description Also known as
English
On sums of squares of primes and a \(k\)-th power of prime
scientific article

    Statements

    On sums of squares of primes and a \(k\)-th power of prime (English)
    0 references
    0 references
    0 references
    22 January 2019
    0 references
    This paper considers the three quantities related to ``exceptional sets'': -- $E_3(k,N)$ is the number of integers below $N$ and satisfying the obvious congruences conditions that are *not* of the form $p_1^2+p_2^2+p_3^k$. -- $E_4(k,N)$ is the number of integers below $N$ and satisfying the obvious congruences conditions that are *not* of the form $p_1^2+p_2^2+p_3^2+p_4^k$. -- $E(k,N)$ is the number of integers below $N$ and satisfying the obvious congruences conditions that are *not* of the form $p_1+p_2^2+p_3^k$. It is proved by using the circle method that $E_3(k,N)\ll_\varepsilon N^{1-\frac{1}{\theta(k)}+\varepsilon}$ for any positive $\varepsilon$ and where $$\theta(k)= \left\{\begin{aligned} 12\text{ when }k=3,\\ \min(2^{k/2+2}, k^2+2k)\text{ when }k\ge4\text{ even},\\ \min(3\cdot2^{(k+1)/2}, 8\lceil \tfrac18(k+1)^2\rceil)\text{ when }k\ge4\text{ odd}.\\ \end{aligned}\right.$$ This improves on several results, the most recent of which being due to \textit{J. Brüdern} [in: From arithmetic to zeta-functions. Number theory in memory of Wolfgang Schwarz. Cham: Springer. 57--81 (2016; Zbl 1407.11117)]. The estimate $E_4(k,N)\ll_\varepsilon N^{\frac12-\frac{1}{\theta(k)}+\varepsilon}$ is also proved. This improves on work of \textit{T. Li} [Rocky Mt. J. Math. 42, No. 1, 201--222 (2012; Zbl 1243.11099)]. The new ingredients comes from \textit{L. Zhao} [Proc. Lond. Math. Soc. (3) 108, No. 6, 1593--1622 (2014; Zbl 1370.11116)] and \textit{J. Bourgain} [Proc. Steklov Inst. Math. 296, 30--40 (2017; Zbl 1371.11138); translation in Tr. Mat. Inst. Steklova 296, 36--46 (2017)]. Finally, the authors prove the estimate $E(k,N)\ll N^{1-\frac{1}{\theta^*(k)}+\varepsilon}$, where $$\theta^*(k)= \left\{\begin{aligned} 15\text{ when }k=3,\\ 36\text{ when }k=4,\\ k(k^2+k+2)/2\text{ when }k\ge5.\\ \end{aligned}\right.$$ This improves on the work of \textit{T. Li} [Rocky Mt. J. Math. 42, No. 1, 201--222 (2012; Zbl 1243.11099)]. The additional ingredient comes from \textit{L. Zhao} [Proc. Lond. Math. Soc. (3) 108, No. 6, 1593--1622 (2014; Zbl 1370.11116)].
    0 references
    0 references
    Waring-Goldbach problem
    0 references
    circle method
    0 references
    exceptional set
    0 references
    0 references