Plancherel-Pólya’s Type of Instability in Vibration System with Multiple Frozen Arguments

Lung-Hui Chen, Chung-Tsun Shieh · 2025

Read Plancherel-Pólya’s Type of Instability in Vibration System with Multiple Frozen Arguments by Lung-Hui Chen, Chung-Tsun Shieh (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-2508-11101.txt.

Introduction

Vibrating strings and rods are described by Sturm-Liouville equations, and hearing their tones to deduce their makeup is a classic inverse problem. Lung-Hui Chen and Chung-Tsun Shieh study a twist on that problem: the equation's argument is frozen at several fixed points inside the interval. The standing waves then behave like slightly disturbed sine or cosine waves.

The disturbance can be written as a Fourier transform of some function, which is where trouble can enter. The authors show the spectral shift obeys an inequality of Plancherel-Pólya type, a bound borrowed from the theory of entire functions. They connect that bound to how far the hidden potential can move in the square-integrable norm. The result maps a precise kind of instability for these frozen-argument systems.

Catalog record

CREATORLung-Hui Chen, Chung-Tsun Shieh
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE29 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2508-11101

Browse the full library · About the Tower of Babel

Full text

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

Plancherel-Pólya’s Type of Instability in Vibration System with Multiple Frozen Arguments Lung-Hui Chen, Chung-Tsun Shieh Abstract We discuss the problem of the inverse spectral problem of Sturm-Liouville operator with multiple frozen arguments at { a 1 , a 2 , … , a N } \{a_{1},a_{2},\ldots,a_{N}\} in ( 0 , π ) (0,\pi) . One may consider the characteristic functions as perturbation of sine or of cosine functions depending on the boundary problem prescribed. However, such perturbation is represented in the form of Fourier transform of certain function which may or may not bring in Riesz basis theory and classical perturbation theory in functional analysis. We shall demonstrate the spectral perturbation in Plancherel-Pólya’s type of inequality and connect to perturbation of related potential functions in L 2 L^{2} -functional norm. MSC: 34A55/34K29/65L03. Keywords: Sturm–Liouville operator; frozen argument; inverse spectral problem; sine type function; instability; interpolation theory. 1 Introduction In this paper, we study the instability problem on recovering the real-valued potential function q(x)\in L^{2}(0,\pi) from the spectrum of boundary value problem L=L(q(x),a_{1},a_{2},\ldots,a_{N},\alpha,\beta) in form of \displaystyle\left\{\begin{array}[]{ll}ly:=-y^{\prime\prime}(x)+q(x)\sum_{i=1}^{N}y(a_{i})=\lambda y(x),\,0<x<\pi;\\ y^{(\alpha)}(0)=y^{(\beta)}(\pi)=0,\end{array}\right. in which \lambda=\rho^{2},\,\rho\in\mathbb{C}, is the spectral parameter, and \alpha , \beta\in\{0,1\} . In system ( 1 ), we assume that 0<a_{1}<a_{2}<\cdots<a_{N}<\pi . The operator l is called the Sturm-Liouville type of operator with N frozen arguments that is compatible with either Dirichlet or Neumann boundary condition on the end points of [0,\pi] . In mathematical or engineering literature, the system ( 1 ) is called one kind of loaded differential equations. The system simulates various experiments in applied physics, mathematical physics, and many engineering projects focusing on the control and measurement of certain vibrations, magnetic shielding in certain oscillatory systems [ 10 , 11 , 17 , 18 ] . In oscillatory systems, one often needs to place certain measurement instruments or sensors to monitor the on-site physical states of the vibration systems which justifies the physical senses of frozen points in ( 1 ). The inverse spectral problems for Sturm-Liouville operators with frozen arguments were previously studied in [ 1 , 2 , 3 , 4 , 5 , 6 , 7 , 8 , 9 , 12 , 13 , 14 , 19 ] to represent a limited selection, and is highly-pursued by many mathematicians and engineers. Among the rapidly growing researches as aforementioned, there are concerns on the local stability problem about how the spectral data of system ( 1 ) is related to the potential function q(x) or the other classes of information provided by the characteristic functions. Let us refer to [ 4 , 5 , 6 , 7 , 13 , 14 ] to list merely a few recent results. The problem seems to be ill-posed on many respects of the system ( 1 ), so we investigate the instability of problem ( 1 ) via the point of view of interpolation theory in Fourier analysis, and Plancherel-Pólya’s type of inequality of the class of entire function of finite type plays a role. 2 Preliminaries We begin by referencing several identities from [ 4 , 5 , 19 ] , specifically their identities (25), (26), (27), and (28) in Shieh and Tsai [ 19 ] . The eigenvalues of differential system ( 1 ) are the zeros of the characteristic functions referring to [ 12 , 19 ] . Here are the characteristic functions all we need for system ( 1 ) for \alpha,\beta\in\{0,1\} : \displaystyle\hskip-6.0pt\Delta_{N}^{(0,0)}(\lambda)=\begin{vmatrix}\frac{\sin\rho a_{1}}{\rho}&\hskip 5.0pt\int_{0}^{a_{1}}q(t)\frac{\sin\rho(a_{1}-t)}{\rho}dt-1&1&1&1&\dots&1&1\\ \frac{\sin\rho a_{2}}{\rho}&\int_{0}^{a_{2}}q(t)\frac{\sin\rho(a_{2}-t)}{\rho}dt&-1&0&0&\dots&0&0\\ \frac{\sin\rho a_{3}}{\rho}&\int_{0}^{a_{3}}q(t)\frac{\sin\rho(a_{3}-t)}{\rho}dt&0&-1&0&\dots&0&0\\ \vdots&\vdots&\vdots&\vdots&\vdots&\ddots&\vdots&\vdots\\ \frac{\sin\rho a_{N}}{\rho}&\int_{0}^{a_{N}}q(t)\frac{\sin\rho(a_{N}-t)}{\rho}dt&0&0&0&\dots&0&-1\\ \frac{\sin\rho\pi}{\rho}&\int_{0}^{\pi}q(t)\frac{\sin\rho(\pi-t)}{\rho}dt&0&0&0&\dots&0&0\end{vmatrix}, (2.1) \displaystyle\hskip-5.0pt\Delta_{N}^{(0,1)}(\lambda)=\begin{vmatrix}\frac{\sin\rho a_{1}}{\rho}&\hskip 3.0pt\int_{0}^{a_{1}}q(t)\frac{\sin\rho(a_{1}-t)}{\rho}dt-1&1&1&1&\dots&1&1\\ \frac{\sin\rho a_{2}}{\rho}&\int_{0}^{a_{2}}q(t)\frac{\sin\rho(a_{2}-t)}{\rho}dt&-1&0&0&\dots&0&0\\ \frac{\sin\rho a_{3}}{\rho}&\int_{0}^{a_{3}}q(t)\frac{\sin\rho(a_{3}-t)}{\rho}dt&0&-1&0&\dots&0&0\\ \vdots&\vdots&\vdots&\vdots&\vdots&\ddots&\vdots&\vdots\\ \frac{\sin\rho a_{N}}{\rho}&\int_{0}^{a_{N}}q(t)\frac{\sin\rho(a_{N}-t)}{\rho}dt&0&0&0&\dots&0&-1\\ \cos\rho\pi&\int_{0}^{\pi}q(t)\cos\rho(\pi-t)dt&0&0&0&\dots&0&0\end{vmatrix}, (2.2) \displaystyle\hskip-4.0pt\Delta_{N}^{(1,0)}(\lambda)=\begin{vmatrix}\cos\rho a_{1}&\hskip-2.0pt\int_{0}^{a_{1}}q(t)\frac{\sin\rho(a_{1}-t)}{\rho}dt-1&1&1&1&\dots&1&1\\ \cos\rho a_{2}&\int_{0}^{a_{2}}q(t)\frac{\sin\rho(a_{2}-t)}{\rho}dt&-1&0&0&\dots&0&0\\ \cos\rho a_{3}&\int_{0}^{a_{3}}q(t)\frac{\sin\rho(a_{3}-t)}{\rho}dt&0&-1&0&\dots&0&0\\ \vdots&\vdots&\vdots&\vdots&\vdots&\ddots&\vdots&\vdots\\ \cos\rho a_{N}&\int_{0}^{a_{N}}q(t)\frac{\sin\rho(a_{N}-t)}{\rho}dt&0&0&0&\dots&0&-1\\ \cos\rho\pi&\int_{0}^{\pi}q(t)\frac{\sin\rho(\pi-t)}{\rho}dt&0&0&0&\dots&0&0\end{vmatrix}, (2.3) and \displaystyle\hskip-1.0pt\Delta_{N}^{(1,1)}(\lambda)=\hskip-2.0pt\begin{vmatrix}\cos\rho a_{1}&\hskip-4.0pt\int_{0}^{a_{1}}q(t)\frac{\sin\rho(a_{1}-t)}{\rho}dt-\hskip-2.0pt1&1&1&1&\hskip-2.0pt\dots&1&1\\ \cos\rho a_{2}&\int_{0}^{a_{2}}q(t)\frac{\sin\rho(a_{2}-t)}{\rho}dt&-1&0&0&\hskip-2.0pt\dots&0&0\\ \cos\rho a_{3}&\int_{0}^{a_{3}}q(t)\frac{\sin\rho(a_{3}-t)}{\rho}dt&0&\hskip-2.0pt-1&0&\hskip-3.0pt\dots&0&0\\ \vdots&\vdots&\vdots&\vdots&\vdots&\hskip-3.0pt\ddots&\vdots&\vdots\\ \cos\rho a_{N}&\int_{0}^{a_{N}}q(t)\frac{\sin\rho(a_{N}-t)}{\rho}dt&0&0&0&\hskip-3.0pt\dots&0&-1\\ -\rho\sin\rho\pi&\hskip-3.0pt\int_{0}^{\pi}q(t)\cos\rho(\pi-t)dt&0&0&0&\hskip-3.0pt\dots&0&0\end{vmatrix}. (2.4) Then, we deduce from ( 2.1 ), that is for the case \alpha=\beta=0 , the following integral identity \displaystyle\Delta^{(0,0)}_{N}(\lambda) \displaystyle\vskip-2.0pt=\vskip-2.0pt \displaystyle\frac{\sin\rho\pi}{\rho}-\frac{\sin\rho\pi}{\rho^{2}}\Big\{\sum_{i=1}^{N}\int_{0}^{a_{i}}\sin{\rho(a_{i}-t)}q(t)dt\Big\} (2.5) \displaystyle+\frac{\sum_{i=1}^{N}\sin\rho a_{i}}{\rho^{2}}\int_{0}^{\pi}\sin{\rho(\pi-t)}q(t)dt. Since the characteristic function depends on the potential function q , we equivalently write \displaystyle\Delta^{(0,0)}_{N}(q;\rho) \displaystyle\vskip-2.0pt=\vskip-2.0pt \displaystyle\frac{\sin\rho\pi}{\rho}-\frac{\sin\rho\pi}{\rho^{2}}\Big\{\sum_{i=1}^{N}\int_{0}^{a_{i}}\sin{(\rho t)}q(a_{i}-t)dt\Big\} (2.6) \displaystyle+\frac{\sum_{i=1}^{N}\sin\rho a_{i}}{\rho^{2}}\int_{0}^{\pi}\sin{(\rho t)}q(\pi-t)dt, and the other cases have similar expressions. Therefore, one shall only demonstrate the proof for the case (\alpha,\beta)=(0,0) . Furthermore, let q^{1},\,q^{2}\in L^{2}(0,\pi) be potentials parametrizing the system ( 1 ) with the corresponding representation function \displaystyle G^{j}(\rho) \displaystyle:= \displaystyle\Delta^{(0,0)}_{N}(q^{\,j};\rho),\,j=1,2, (2.7) \displaystyle\widehat{q}\,(x) \displaystyle:= \displaystyle q^{1}(x)-q^{2}(x), and \displaystyle\hskip-2.0pt\widehat{G}(\rho) \displaystyle:= \displaystyle\Delta^{(0,0)}_{N}(q^{1};\rho)-\Delta^{(0,0)}_{N}(q^{2};\rho) (2.8) \displaystyle= \displaystyle-\frac{\sin\rho\pi}{\rho^{2}}\Big\{\sum_{i=1}^{N}\int_{0}^{a_{i}}\sin(\rho t)\,\widehat{q}(a_{i}-t)dt\Big\} \displaystyle+\frac{\sum_{i=1}^{N}\sin\rho a_{i}}{\rho^{2}}\int_{0}^{\pi}\sin(\rho t)\,\widehat{q}(\pi-t)dt. (2.9) Let us denote the zero set of ( 2.9 ) as \mathcal{Z}(\widehat{G}):=\{\alpha_{1},\alpha_{2},\ldots\}, which includes the eigenvalue set of system ( 1 ) when (\alpha,\beta)=(0,0) . We also rephrase ( 2.9 ) to be \displaystyle\widehat{G}(\rho) \displaystyle= \displaystyle-\frac{\sin\rho\pi}{\rho^{2}}\Im\Big\{\int_{-\pi}^{\pi}e^{-i\rho t}\{\sum_{i=1}^{N}\,\chi_{[0,a_{i}]}(t)\,\widehat{q}(a_{i}-t)\}dt\Big\} (2.10) \displaystyle+\frac{\sum_{i=1}^{N}\sin\rho a_{i}}{\rho^{2}}\Im\Big\{\int_{-\pi}^{\pi}e^{-i\rho t}\chi_{[0,\pi]}(t)\,\widehat{q}(\pi-t)dt\Big\}, in which the cut-off function on [a,b] \displaystyle\chi_{[a,b]}(t):=\left\{\begin{array}[]{ll}1,\,\mbox{ for }t\in[a,b];\\ 0,\,\mbox{ otherwise}.\end{array}\right. Now, we apply the Paley-Wiener theorem [ 16 , p. 151] , for each function \psi(t)\in L^{2}(-\pi,\pi) , there exists the representation F(\rho)=\frac{1}{2\pi}\int_{-\pi}^{\pi}e^{-i\rho t}\psi(t)dt,\mbox{ where }F(\rho)\in L^{2}_{\pi}(-\infty,\infty), (2.14) which denotes entire functions that is L^{2} -integrable on the real axis and at most of finite type \pi . We refer the definitions and details to Appendix. Therefore, we deduce \displaystyle\|\widehat{G}(\rho)\|_{L^{2}(-\infty,\infty)} \displaystyle\leq \displaystyle\|\frac{\sin\rho\pi}{\rho^{2}}\|_{L^{2}(-\infty,\infty)}\|\mathcal{F}(\sum_{i=1}^{N}\chi_{[0,a_{i}]}\widehat{q}(a_{i}-t))\|_{L^{2}(-\infty,\infty)} (2.15) \displaystyle\hskip-5.0pt+\|\frac{\sum_{i=1}^{N}\sin\rho a_{i}}{\rho^{2}}\|_{L^{2}(-\infty,\infty)}\|\mathcal{F}(\chi_{[0,\pi]}(t)\widehat{q}(\pi-t))\|_{L^{2}(-\infty,\infty)}. Using Parseval’s identity to ( 2.15 ), we obtain \displaystyle\hskip-2.0pt\|\widehat{G}(\rho)\|_{L^{2}(-\infty,\infty)} \displaystyle\leq \displaystyle\frac{1}{\sqrt{2\pi}}\,\|\frac{\sin\rho\pi}{\rho^{2}}\|_{L^{2}(-\infty,\infty)}\sum_{i=1}^{N}\|\chi_{[0,a_{i}]}(t)\widehat{q}(a_{i}-t)\|_{L^{2}(-\pi,\pi)} (2.16) \displaystyle\hskip-15.0pt+\frac{1}{\sqrt{2\pi}}\,\|\frac{\sum_{i=1}^{N}\sin\rho a_{i}}{\rho^{2}}\|_{L^{2}(-\infty,\infty)}\|\chi_{[0,\pi]}(t)\widehat{q}(\pi-t)\|_{L^{2}(-\pi,\pi)}, and \displaystyle\|\widehat{G}(\rho)\|_{L^{2}(-\infty,\infty)} \displaystyle\leq \displaystyle\frac{N}{\sqrt{2\pi}}\,\|\frac{\sin\rho\pi}{\rho^{2}}\|_{L^{2}(-\infty,\infty)}\|\{\hat{c}_{k}\}_{k\in\mathbb{Z}}\|_{l^{2}} (2.17) \displaystyle+\frac{1}{\sqrt{2\pi}}\,\|\frac{\sum_{i=1}^{N}\sin\rho a_{i}}{\rho^{2}}\|_{L^{2}(-\infty,\infty)}\|\{\hat{c}_{k}\}_{k\in\mathbb{Z}}\|_{l^{2}}, in which the set \{\hat{c}_{k}\}_{k\in\mathbb{Z}} are the Fourier coefficients of \mathcal{F}(\widehat{q}\,)(\rho) and \mathcal{F}(\widehat{q}\,)(k)=\hat{c}_{k} . For \rho\in\mathbb{C} , we note \displaystyle\sin{\rho\pi}=O(e^{|\Im\rho|\pi}); \displaystyle\hskip-21.0pt\sum_{i=1}^{N}\,\sin\rho a_{i}=O(e^{|\Im\rho|a_{N}}), which implies that \sin{\rho\pi} is an entire function of type \pi and \sum_{i=1}^{N}\sin\rho a_{i} is of type a_{N} that are non-trivial entire functions with solid set with zero density. We refer the details to Appendix. Proposition 2.1 (Cartwright) . The entire function \widehat{G}(\rho) is of finite type and can be represented as \widehat{G}(\rho)=c\rho^{m}\lim_{R\rightarrow\infty}\prod_{|\alpha_{k}|\leq R}\Big(1-\frac{\rho}{\alpha_{k}}\Big),\,k\in\mathbb{Z}, (2.18) where \alpha_{k}\in\mathcal{Z}(\widehat{G}) , c is a constant, and \rho^{2}\widehat{G}(\rho)=o(1) for \Re\rho\rightarrow\pm\infty . Proof. We observe that \widehat{G}(\rho) is bounded over the real axis with non-trivial q\in L^{2}(0,\pi) . As a result of Cartwright’s theory stated as ( 4.4 ) in Theorem 4.8 in Appendix, we begin that \widehat{G}(\rho)=c\,\rho^{m}e^{i\kappa\rho}\lim_{R\rightarrow\infty}\prod_{|\alpha_{k}|\leq R}\Big(1-\frac{\rho}{\alpha_{k}}\Big),\,\alpha_{k}\in\mathcal{Z}(\widehat{G}), (2.19) where c,m,\kappa are constants and \kappa is real, and \alpha_{k}\in\mathcal{Z}(\widehat{G}) . Now we consider \rho\rightarrow\pm\infty on real axis, we deduce that \widehat{G}(\rho)\rightarrow 0, using Riemann-Lebesgue Lemma to ( 2.9 ), and then deduce that the real constant \kappa=0 . Then, we obtain \widehat{G}(\rho)=c\,\rho^{m}\lim_{R\rightarrow\infty}\prod_{|\alpha_{k}|\leq R}\Big(1-\frac{\rho}{\alpha_{k}}\Big), which is bounded on \Im\rho=0 . Hence, m\leq 0 . Thus, we deduce that \widehat{G}(\rho)=c\lim_{R\rightarrow\infty}\prod_{|\alpha_{k}|\leq R}\Big(1-\frac{\rho}{\alpha_{k}}\Big),\,\alpha_{k}\in\mathcal{Z}(\widehat{G}). (2.20) To retrieve the constant c , we let \rho\rightarrow 0 in equation ( 2.9 ) \displaystyle\widehat{G}(\rho) \displaystyle\vskip-5.0pt=\vskip-5.0pt \displaystyle-\frac{\sin\rho\pi}{\rho}\Big\{\sum_{i=1}^{N}\int_{0}^{a_{i}}\frac{\sin{(\rho t)}}{\rho}\,\widehat{q}(a_{i}-t)dt\Big\} (2.21) \displaystyle+\,\frac{\sum_{i=1}^{N}\sin\rho a_{i}}{\rho}\int_{0}^{\pi}\frac{\sin(\rho t)}{\rho}\,\widehat{q}(\pi-t)dt. Using the formula \int_{0}^{\alpha}u(t)\sin\rho tdt=o(e^{|\Im\rho|\alpha}),\,\alpha\in(0,\infty), (2.22) we deduce from ( 2.21 ) that \widehat{G}(\rho)=e^{|\Im\rho|(\pi+a_{N})}o(\frac{1}{\rho^{2}}),\,|\rho|\gg 0,\,\rho\in\mathbb{C}. (2.23) The proportion is thus proven. ∎ 3 Results Let us denote by L^{p}_{\sigma} , 1\leq p<\infty , the space of entire functions of exponential type \leq\sigma that belong to the space L^{p}(-\infty,\infty) . Let us apply the Plancherel-Pólya theorem, and refer to [ 16 , p. 149] as in Appendix. Theorem 3.1 . Let q^{\,j}\in L^{2}(0,\pi) be the potential parametrizing the system ( 1 ) with Fourier coefficients \{c^{\,j}_{k}\}=\mathcal{F}(q^{\,j})(k) , j=1,2 , k\in\mathbb{Z} , and we write \widehat{q}(x)=q^{1}(x)-q^{2}(x). Then, for some C>0 and some h\gg 0 , one has (h+1)(a_{N}+|\mbox{c.h. supp }\widehat{q}(t)|)\geq-\ln\frac{|C|\|\{\hat{c}_{k}\}_{k\in\mathbb{Z}}\|_{l^{2}}^{2}}{|\widehat{G}(x\pm ih)|^{2}}, in which |\mbox{c.h. supp }\widehat{q}(t)| means the measure of the convex hull of the effective support of \widehat{q}(t) . Proof. Let

Read the complete text