On continuous and discrete Hardy inequalities (Q516575)

The authors furnish a series of \textbf{d}iscrete and \textbf{c}ontinuous \textbf{H}ardy \textbf{i}nequalities (\textbf{dHi} and \textbf{cHi} for short), principally we state significant results of each section. Accordingly, the second section focuses on the continuous case, the authors state the generalized \textbf{cHi} with multiple singularities, i.e., \[ (*):\;\displaystyle\int_{\mathbb R^d}|\nabla u(x)|^2dx\geq\frac{(d-2)^2}{4} \int_{\mathbb R^d}\frac{G(x)}{[w(x)]^2}|u(x)|^2dx \] with \(G(x)=\displaystyle\sum_{j=1}^d\left(\sum_{k=1}^n\frac{x_j-a_{k_j}}{|x-a_k|^d}\right)^2\), \(u(x)\) is a smooth decaying function, \(a_k=(a_{k_j})_{1\leq j\leq d}\in\mathbb R^d\) for \(k\in\{1,\ldots,n\}\), and \(\displaystyle w=\sum_{k=1}^n\frac{1}{|x-a_k|^{d-2}}\), and \(d\geq 3\) (Theorem 2.1). The proof is based on using technical calculus. Then by assuming that \(u(x)=0\) for \(x\) belonging to \( \partial S_n\), the boundary of \(S_n=\left\{x\in\mathbb R^2:\displaystyle\prod_{k=1}^n|x-a_k|\leq 1\right\}\), the authors obtain the same inequality as \((*)\) on \(\mathbb R^2\setminus S_n\) with the coefficient \(\displaystyle\frac{1}{4}\) instead of \(\displaystyle\frac{(d-2)^2}{4}\) (Theorem 2.2). The third section deals with \textbf{dHi} on \(\mathbb Z^d\) for \(d\geq 3\). Precisely, for \(f\) a function defined on \(\mathbb Z^d\), they show \[ \displaystyle \sum_{n\in\mathbb Z^d}\frac{|f(n)|^2}{|n|^2}\leq C \sum_{n\in\mathbb Z^d}\sum_{j=1}^d|f(n)-f(n-{1}_j)|^2, \] where \({1}_j\) is the unit vector in the direction \(j\) and \(C\) is a constant bounded by an explicit function of \(d\) (Theorem 3.1). As regards to the proof, the authors use, essentially, the Parseval identity and \textbf{cHi} on the torus (Lemma 3.2). The fourth section concentrates on \textbf{dHi} on \(\mathbb Z^2\), and the authors state that there is a positive constant \(K\) such that \[ \displaystyle\sum_{x\in\mathbb Z^2,|x|\geq 2}\frac{|f(x)|^2}{|x|^2(\ln|x|)^2}\leq K\sum_{(x,y)\in A}|f(x)-f(y)|^2, \] such that \(A=\{(x,y)\in\mathbb Z^2:x\sim y\}\), where \(x\sim y\) means that there is an edge between \(x\) and \(y\), \(f\) is a function with compact support and \(f(x)=0\) if \(|x|\leq 1\) (Theorem 4.1). The proof is a little bit long and based on using new apparatuses as discrete polar coordinates on \(\mathbb Z^2\) and discrete Poincaré-Friedrichs type inequality (Lemmas 4.2 and 4.4).