On the density version of quadratic Waring's problem and the quadratic Waring--Goldbach problem
Zi Li Lim
TL;DR
This work establishes sharp density-type results for quadratic Waring and quadratic Waring–Goldbach problems in modular and integer settings. It develops a transference framework, anchored by restriction estimates for squares in cyclic groups and detailed sumset analyses, to prove that a dense subset of squares in $(\mathbb{Z}/W)^{(2)}$ densely generates all residues via sums of $s$ terms when $s\ge 5$ and $W$ avoids small primes. For integers, it provides improvements on density thresholds in the Waring–Goldbach context, including an explicit local-to-global scheme with thresholds $D_s$ and $d_s$ and a congruence condition; the results improve prior density bounds in several regimes and illuminate the role of local moduli. The paper combines analytic tools (restriction theory, Gauss sums), additive combinatorics (sumsets, Green–Ruzsa), and a combinatorial lemma with computer-assisted base cases to bridge local and global results, including extensions to small moduli via a refined $W$-trick. Overall, it advances the understanding of density phenomena in quadratic representations and connects modular and integer Waring-type problems through a flexible transference paradigm.
Abstract
We prove a sharp density theorem for quadratic Waring's problem over cyclic groups, when the number of variables is at least $5$. Also, we obtain some new improvements on the density version of the quadratic Waring--Goldbach problem over integers.
