Remarks on Redheffer’s inequality

Nagi Suzuki, Shingo Takeuchi · 2025

Read Remarks on Redheffer’s inequality by Nagi Suzuki, Shingo Takeuchi (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2510-20841.txt.

Introduction

Nagi Suzuki and Shingo Takeuchi give remarks on Redheffer's inequality. An inequality states that one amount is larger than another. They use this result and its wider forms to study one value. That value is the first eigenvalue of the p-Laplacian. An eigenvalue is a special value tied to how a system behaves. They track how it changes as the value p changes. They extend a Redheffer type result to a broader class too. That result concerns the generalized trigonometric function, a broad form of sine and cosine.

Catalog record

CREATORNagi Suzuki, Shingo Takeuchi
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE17 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2510-20841

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Remarks on Redheffer’s inequality Thanks: The work of S. Takeuchi was supported by JSPS KAKENHI Grant Number 22K03392. Nagi Suzuki, Shingo Takeuchi Abstract . Redheffer’s inequality and its generalization are applied to the study of behavior and estimates of the first eigenvalue of p p -Laplacian with respect to p p . Furthermore, a Redheffer-type inequality for the generalized trigonometric function is extended to a broader class. 1991 Mathematics Subject Classification 34L15, 26D05, 33E05 1. Introduction Redheffer’s inequality (1.1) \frac{\pi^{2}-x^{2}}{\pi^{2}+x^{2}}\leq\frac{\sin{x}}{x} was proposed by R. Redheffer in 1968 as an advanced problem in Amer. Math. Monthly [ 10 ] *No. 5642. In another issue of the journal [ 15 ] *No. 5642, the inequality was titled A delightful inequality and J.P. Williams gave a proof relying on the infinite product representation of the sine function. Since then, Redheffer’s inequality has been widely studied in the area of inequality (see [ 18 , 20 , 19 , 13 , 16 , 4 , 14 ] and the references given there). Research on Redheffer’s inequality primarily consists of work aimed at improving its precision as an inequality and discovering similar inequalities involving trigonometric functions, hyperbolic functions, or special functions. See for instance [ 1 , 19 , 20 , 14 ] . To the authors’ knowledge, there have been few applications of this delightful inequality. The purpose of this paper is to apply Redheffer’s inequality and its generalization by Zhu and Sun [ 19 ] to the study of eigenvalue problems for differential equations, specifically to the analysis of behavior and estimates of the first eigenvalue of p -Laplacian with respect to p . As a result, we can provide an alternative proof for the monotonicity of the first eigenvalue shown by Kajikiya, Tanaka and Tanaka [ 12 ] , while also refining the estimate for the first eigenvalue obtained by Kajikiya and Takeuchi [ 11 ] . Furthermore, we extend the Redheffer-type inequality for generalized cosine function, previously discovered by Ozawa and Takeuchi [ 14 ] , to a broader class. 2. Preliminaries In this section, we define generalized trigonometric functions and generalized \pi necessary for stating the results in the next section. Let p,q\in(1,\infty) and F_{p,q}(x):=\int_{0}^{x}\frac{dt}{(1-t^{q})^{1/p}},\quad x\in[0,1]. We will denote by \sin_{p,q} the inverse function of F_{p,q} , i.e., \sin_{p,q}{x}:=F_{p,q}^{-1}(x). Clearly, \sin_{p,q}{x} is an increasing function in [0,\pi_{p,q}/2] to [0,1] , where \pi_{p,q}:=2F_{p,q}(1)=2\int_{0}^{1}\frac{dt}{(1-t^{q})^{1/p}}=\frac{2}{q}B\left(1-\frac{1}{p},\frac{1}{q}\right), where B is the beta function. We extend \sin_{p,q}{x} to (\pi_{p,q}/2,\pi_{p,q}] by \sin_{p,q}{(\pi_{p,q}-x)} and to the whole real line \mathbb{R} as the odd \pi_{p,q} -antiperiodic continuation of the function. Since \sin_{p,q}{x}\in C^{1}(\mathbb{R}) , we also define \cos_{p,q}{x} by \cos_{p,q}{x}:=(\sin_{p,q}{x})^{\prime} , where {}^{\prime}:=d/dx . Then, it follows that |\cos_{p,q}{x}|^{p}+|\sin_{p,q}{x}|^{q}=1. In case (p,q)=(2,2) , it is obvious that \sin_{p,q}{x},\ \cos_{p,q}{x} and \pi_{p,q} are reduced to the ordinary \sin{x},\ \cos{x} and \pi , respectively. This is the reason why these functions and the constant are called generalized trigonometric functions and generalized \pi (by parameter (p,q) ), respectively. We denote \sin_{p,p}{x},\ \cos_{p,q}{x} and \pi_{p,p} by \sin_{p}{x},\ \cos_{p}{x} and \pi_{p} , respectively. In particular, it is noted that \pi_{p}=\frac{2}{p}B\left(1-\frac{1}{p},\frac{1}{p}\right)=\frac{2\pi}{p\sin{(\pi/p)}}. We conclude this section by explaining the utility of generalized trigonometric functions and generalized \pi . Let \lambda(p) denote the first eigenvalue of p -Laplacian with zero Dirichlet boundary conditions on the interval [-1,1] , and let \varphi denote the corresponding eigenfunction with \max_{-1\leq x\leq 1}\varphi(x)=1 . That is, \lambda(p) and \varphi satisfy -(|\varphi^{\prime}|^{p-2}\varphi^{\prime})^{\prime}=\lambda(p)|\varphi|^{p-2}\varphi,\quad\varphi(-1)=\varphi(1)=0. Then, it is known that \lambda(p) and \varphi can be expressed as follows (see, for example, [ 8 , 7 ] ): (2.1) \displaystyle\lambda(p)=(p-1)\left(\frac{\pi_{p}}{2}\right)^{p}=(p-1)\left(\frac{\pi}{p\sin{(\pi/p)}}\right)^{p}, \displaystyle\varphi(x)=\sin_{p}{\left(\frac{\pi_{p}}{2}(x+1)\right)}. Using \lambda(p) and \varphi , all solutions (\lambda,u) to the eigenvalue problem -(|u^{\prime}|^{p-2}u^{\prime})^{\prime}=\lambda|u|^{p-2}u,\quad u(a)=u(b)=0 can be fully written down as follows: \lambda_{k}=\left(\frac{2k}{b-a}\right)^{p}\lambda(p),\quad u_{k}(x)=C\varphi\left(\frac{2k(x-a)}{b-a}-1\right), where k=1,2,3,\ldots and C is an arbitrary constant. Thus, \lambda(p) and \varphi provide us with a representation of the solutions to the eigenvalue problem for p -Laplacian. 3. Main results This section presents the three main results of this paper. The first two are applications of Redheffer’s inequality to the analysis of the first eigenvalue \lambda(p) of p -Laplacian, while the remaining one is an extension of a known Redheffer-type inequality. 3.1. Alternative proof of the monotonicity of \lambda(p) Kajikiya, Tanaka, and Tanaka [ 12 ] show that the behavior of the first eigenvalue of the p -Laplacian on the interval [-L,L] with respect to p changes significantly at the threshold L=1 . Here, we present their result for L=1 . The first eigenvalue of the p -Laplacian in this case is the \lambda(p) given by ( 2.1 ). Proposition 3.1 ( [ 12 ] ) . \lambda(p) is strictly increasing with respect to p in (1,\infty) . Below, we provide an alternative proof of Proposition 3.1 using Redheffer’s inequality ( 1.1 ). First, let us explain their proof idea in [ 12 ] . They prove Proposition 3.1 using standard differential calculus. Ultimately, the proof reduces to the problem of determining the sign of a certain function. Although they do not explicitly state it as a lemma, we will present it here in the form of a lemma for convenience. Lemma 3.2 . Let k(x):=x^{2}(\pi-x)^{2}-(\pi^{2}-\pi x+x^{2})\sin^{2}{x}. Then, \lambda^{\prime}(p)>0 for all p\in(1,\infty) if k(x)<0 for all x\in(0,\pi) . By this lemma, to prove Proposition 3.1 , it suffices to show that k(x)<0 . We briefly review their (skillful) proof in [ 12 ] that k(x)<0 . By the symmetry of k(x) , we may assume 0<x<\pi/2 . Then, since \sin x>x-x^{3}/6 , k(x)<-\frac{x^{3}}{36}(x^{5}-\pi x^{4}-(12-\pi^{2})x^{3}+12\pi x^{2}-12\pi^{2}x+36\pi). Let the quintic expression on the right-hand side be K(x) . Furthermore, since \displaystyle K^{\prime}(x) \displaystyle=5x^{4}-4\pi x^{3}-3(12-\pi^{2})x^{2}+24\pi x-12\pi^{2} \displaystyle=-(4\pi-5x)x^{3}-3(12-\pi^{2})x^{2}-12\pi(\pi-2x)<0 holds for x\in(0,\pi/2) , we obtain K(x)>K\left(\frac{\pi}{2}\right)=\frac{3}{32}\pi^{5}-\frac{9}{2}\pi^{3}+36\pi=2.25844\cdots>0. Therefore, k(x)<-x^{3}K(x)/36<0 . Thus, Proposition 3.1 follows from Lemma 3.2 . However, in fact, using Redheffer’s inequality ( 1.1 ), we can immediately see that k(x)<0 as follows. Dividing both sides of k(x) by x^{2} , we have \frac{k(x)}{x^{2}}=(\pi-x)^{2}-(\pi^{2}-\pi x+x^{2})\left(\frac{\sin{x}}{x}\right)^{2}. Using ( 1.1 ) to the right-hand side yields \displaystyle\frac{k(x)}{x^{2}} \displaystyle\leq(\pi-x)^{2}-(\pi^{2}-\pi x+x^{2})\left(\frac{\pi^{2}-x^{2}}{\pi^{2}+x^{2}}\right)^{2} \displaystyle=-\frac{\pi x(\pi-x)^{4}}{(\pi^{2}+x^{2})^{2}}<0; hence, k(x)<0 for x\in(0,\pi) . 3.2. Improvement to the estimate of \lambda(p) Regarding the first eigenvalue of the p -Laplacian on an N -dimensional ball, Huang [ 9 ] and Benedikt and Drábek [ 2 ] obtained upper and lower bounds expressed as polynomials in p . Applying their bounds for N=1 yields the following result: Proposition 3.3 ( [ 9 , 2 ] ) . For any p\in(1,\infty) , it holds that p\leq\lambda(p)\leq p+1. When p is close to 1 , the lower bound can be improved (see [ 3 , 5 ] ). However, here we prefer the simplest estimate above. In fact, the upper bound of the inequality in Proposition 3.3 is not optimal even when restricted to linear functions. Indeed, it has recently been improved by Kajikiya and Takeuchi [ 11 ] as follows: Proposition 3.4 ( [ 11 ] ) . For any p\in(1,\infty) , it holds that p<\lambda(p)<p+\frac{\pi^{2}}{6}-1. Moreover, the upper bound is optimal in the following sense: (3.1) \lim_{p\to\infty}(\lambda(p)-p)=\frac{\pi^{2}}{6}-1. Now, we can further adjust the lower bound as follows to accommodate this upper bound. Theorem 3.5 . For any p\in(1,\infty) , it holds that (3.2) p+\frac{\pi^{2}}{6}-1-\frac{\pi^{2}}{6(p+1)}<\lambda(p)<p+\frac{\pi^{2}}{6}-1. Therefore, ( 3.1 ) immediately follows. Below, we will explain the proof of Theorem 3.5 . The proof relies on the following Redheffer-type inequality

Read the complete text