On simultaneous additive equations, IV
Let \(s\) be a positive integer, \(c_i\) and \(d_i\) \((1\leq i\leq s)\) be rational integers, and put \[ F({\mathbf x})=c_1x_1^3+\cdots+c_sx_s^3,\quad G({\mathbf x})=d_1x_1^2+\cdots+d_sx_s^2. \] Suppose that \(G({\mathbf x})\) is indefinite and has at least 5 variables explicit. Suppose also that \(F({\mathbf x})\) has at least 7 variables explicit and that the system \(F({\mathbf x})=G({\mathbf x})=0\) has a non-trivial real solution. (The `trivial' solution is, of course, \({\mathbf x}=(0,\dots,0)\).) Moreover suppose that at least one of the following conditions (i)--(iv) holds: (i) \(s\geq 13\), (ii) \(s\geq 12\) and at least 2 of the \(c_i\) are zero, (iii) at least 6 of the \(d_i\) are zero, (iv) at least 4 of the \(c_i\) are zero. Under these conditions, it is proved in this paper that the simultaneous additive equations \(F({\mathbf x})=G({\mathbf x})=0\) have a non-trivial solution in rational integers. This result refines the author's earlier work [``On simultaneous additive equations. II, J. Reine Angew. Math. 419, 141-198 (1991; Zbl 0721.11011)], where the same conclusion was shown with \(s\geq 14\) in place of (i), and with (ii) absent. [For Parts I and III, see Proc. Lond. Math. Soc. (3) 63, 1-34 (1991; Zbl 0691.10007) respectively Mathematika 37, 85-96 (1990; Zbl 0691.10008)]. The key to this improvement is a mean value theorem established also in this paper. To state it, write \(e(z)\) for \(e^{2\pi iz}\), and define \[ F(\alpha,\beta;P)=\sum_{1\leq x\leq P}e(\alpha x^3+\beta x^2), \quad f(\alpha,\beta; P,R)= \sum_{x\in{\mathcal A}(P,R)} e(\alpha x^3+\beta x^2), \] where \({\mathcal A}(P,R)\) denotes the set of natural numbers \(x\) up to \(P\) such that \(x\) has no prime factors greater than \(R\). Then the author shows that for any \(\varepsilon>0\), there is an \(\eta>0\) such that one has \[ \int_0^1\int_0^1|F(\alpha,\beta;P)^4f(\alpha,\beta;P,P^{\eta})^6|d\alpha d\beta\ll P^{17/3+\varepsilon}, \] where the implicit constant depends at most on \(\varepsilon\) and \(\eta\). In the earlier work cited above, the corresponding result was obtained with the exponent \(35/6\) instead of \(17/3\). The latter progress is due to the efficient differencing method of the author [``On exponential sums over smooth numbers, J. Reine Angew. Math. 488, 79-140 (1997; Zbl 0880.11060)].
- The system of simultaneous equations \(\sum^3_{i=1} x^k_i=\sum^3_{i=1} y^k_i\), \(k=1, 2\) and \(5\) has no non-trivial solutions in integers
- Impossibilité d'un système d'équations indéterminées (question 4514, de \textit{A. Gérardin}).
- On the sums of two cubes
- On some indeterminate equations of third degree.
- On Pairs of Diagonal Cubic Forms
- Problems of Diophantine analysis
- Sur les solutions entières de l'équation \(x^3+y^3=pz^3\).
- On the diophantine equation x^3+y^3+z^3=nxyz
- On perfect powers that are sums of cubes of a five term arithmetic progression
- Propriétés de suites récurrentes (question 4521, de V. G. Tariste (pseudon.)).
- The complete solution of a Diophantine equation involving biquadrates
- The Hasse principle for diagonal forms restricted to lower-degree hypersurfaces
- Multiple exponential sums over smooth numbers
- A problem on semicubical powers
- The paucity problem for simultaneous quadratic and biquadratic equations
- On simultaneous diagonal inequalities, II
- Rational solutions of pairs of diagonal equations, one cubic and one quadratic
- Simultaneous equal sums of three powers
- On simultaneous additive equations II.
- Bounds on 10th moments of (𝑥,𝑥³) for ellipsephic sets
- Rational points on complete intersections of cubic and quadric hypersurfaces over \(\mathbb{F}_q(t)\)
This page was built for publication: On simultaneous additive equations, IV
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4494715)
