Showing 1–21 of 21 results for author: Karabash, I M

Multi-parameter Hopf bifurcations of rimming flows

Authors: Illya M. Karabash, Christina Lienstromberg, Juan J. L. Velázquez

Abstract: In order to investigate the emergence of periodic oscillations of rimming flows, we study analytically the stability of steady states for the model of (Benilov, Kopteva, O'Brien, 2005), which describes the dynamics of a thin fluid film coating the inner wall of a rotating cylinder and includes effects of surface tension, gravity, and hydrostatic pressure. We apply multi-parameter perturbation meth… ▽ More In order to investigate the emergence of periodic oscillations of rimming flows, we study analytically the stability of steady states for the model of (Benilov, Kopteva, O'Brien, 2005), which describes the dynamics of a thin fluid film coating the inner wall of a rotating cylinder and includes effects of surface tension, gravity, and hydrostatic pressure. We apply multi-parameter perturbation methods to eigenvalues of Fréchet derivatives and prove the transition of a pair of conjugate eigenvalues from the stable to the unstable complex half-plane under appropriate variations of parameters. In order to prove rigorously the corresponding branching of periodic solutions from critical equilibria, we extend the multi-parameter Hopf-bifurcation theory to the case of infinite-dimensional dynamical systems. △ Less

Submitted 17 June, 2024; originally announced June 2024.

Comments: 39 pages

MSC Class: 35B32 (primary) 76A20; 37L10; 76U05; 35B10; 35K55; 35K25 (secondary)

Homogenization and nonselfadjoint spectral optimization for dissipative Maxwell eigenproblems

Authors: Matthias Eller, Illya M. Karabash

Abstract: The homogenization of eigenvalues of non-Hermitian Maxwell operators is studied by the H-convergence method. It is assumed that the Maxwell systems are equipped with suitable m-dissipative boundary conditions, namely, with Leontovich or generalized impedance boundary conditions of the form $n \times E = Z [(n \times H )\times n ] $. We show that, for a wide class of impedance operators $Z$, the no… ▽ More The homogenization of eigenvalues of non-Hermitian Maxwell operators is studied by the H-convergence method. It is assumed that the Maxwell systems are equipped with suitable m-dissipative boundary conditions, namely, with Leontovich or generalized impedance boundary conditions of the form $n \times E = Z [(n \times H )\times n ] $. We show that, for a wide class of impedance operators $Z$, the nonzero spectrum of the corresponding Maxwell operator is discrete. To this end, a new continuous embedding theorem for domains of Maxwell operators is obtained. We prove the convergence of eigenvalues to an eigenvalue of a homogenized Maxwell operator under the assumption of the H-convergence of the material tensor-fields. This result is used then to prove the existence of optimizers for eigenvalue optimization problems and the existence of an eigenvalue-free region around zero. As applications, connections with the quantum optics problem of the design of high-Q resonators are discussed, and a new way of the quantification of the unique (and nonunique) continuation property is suggested. △ Less

Submitted 17 February, 2024; v1 submitted 2 January, 2024; originally announced January 2024.

Comments: 31 pages, the discussion on connections with quantum optics and unique continuation is added, in particular, Example 2.2 (ii); 3 references are added; some of the proofs are shortened; typos are corrected

M-dissipative boundary conditions and boundary tuples for Maxwell operators

Authors: Matthias Eller, Illya M. Karabash

Abstract: For Maxwell operators $(E,H) \to (i ε^{-1} \nabla\times H, -i μ^{-1} \nabla \times E)$ in Lipschitz domains, we describe all m-dissipative boundary conditions and apply this result to generalized impedance and Leontovich boundary conditions including the cases of singular, degenerate, and randomized impedance coefficients. To this end we construct Riesz bases in the trace spaces associated with th… ▽ More For Maxwell operators $(E,H) \to (i ε^{-1} \nabla\times H, -i μ^{-1} \nabla \times E)$ in Lipschitz domains, we describe all m-dissipative boundary conditions and apply this result to generalized impedance and Leontovich boundary conditions including the cases of singular, degenerate, and randomized impedance coefficients. To this end we construct Riesz bases in the trace spaces associated with the curl-operator and introduce a modified version of boundary triple adapted for the specifics of Maxwell equations, namely, to the mixed-order duality of the related trace spaces. This provides a translation of the problem to operator-theoretic settings of abstract Maxwell operators. In particular, we show that Calkin reduction operators are naturally connected with Leontovich boundary conditions and provide an abstract version of impedance boundary condition applicable to other types of wave equations. Taking Friedrichs and Krein-von Neumann extensions of related boundary operators, it is possible to associate m-dissipative Maxwell operators to arbitrary non-negative measurable impedance coefficients. △ Less

Submitted 9 October, 2021; originally announced October 2021.

Comments: 38 pages

MSC Class: 35Q61 47B44 35F45 (Primary) 35J56 78A25 58J90 34G10 (Secondary)

Journal ref: Journal of Differential Equations 325 (2022), 82-118

Asymptotics of random resonances generated by a point process of delta-interactions

Authors: Sergio Albeverio, Illya M. Karabash

Abstract: We introduce and study the following model for random resonances: we take a collection of point interactions $Υ_j$ generated by a simple finite point process in the 3-D space and consider the resonances of associated random Schrödinger Hamiltonians $H_Υ= -Δ+ ``\sum \mathfrak{m}(α) δ(x - Υ_j)``$. These resonances are zeroes of a random exponential polynomial, and so form a point process $Σ(H_Υ)$ in… ▽ More We introduce and study the following model for random resonances: we take a collection of point interactions $Υ_j$ generated by a simple finite point process in the 3-D space and consider the resonances of associated random Schrödinger Hamiltonians $H_Υ= -Δ+ ``\sum \mathfrak{m}(α) δ(x - Υ_j)``$. These resonances are zeroes of a random exponential polynomial, and so form a point process $Σ(H_Υ)$ in the complex plane $\mathbb{C}$. We show that the counting function for the set of random resonances $Σ(H_Υ)$ in $\mathbb{C}$-discs with growing radii possesses Weyl-type asymptotics almost surely for a uniform binomial process $Υ$, and obtain an explicit formula for the limiting distribution as $m \to \infty$ of the leading parameter of the asymptotic chain of `most narrow' resonances generated by a sequence of uniform binomial processes $Υ^m$ with $m$ points. We also pose a general question about the limiting behavior of the point process formed by leading parameters of asymptotic sequences of resonances. Our study leads to questions about metric characteristics for the combinatorial geometry of $m$ samples of a random point in the 3-D space and related statistics of extreme values. △ Less

Submitted 3 December, 2019; originally announced December 2019.

Comments: 21 pages

Journal ref: In: Albeverio S., Balslev A., Weder R. (eds.) "Schrödinger Operators, Spectral Analysis and Number Theory. In Memory of Erik Balslev'', Springer Proceedings in Mathematics & Statistics Vol. 348, pp. 7--26, Springer, 2021

Euler-Lagrange equations for full topology optimization of the Q-factor in leaky cavities

Authors: Matthias Eller, Illya M. Karabash

Abstract: We derive Euler-Lagrange equations for the topology optimization of decay rate in 3-d lossy optical cavities. This leads to a new class of time-harmonic differential or integro-differential equations, which can be written as nonlinear Maxwell systems with switching functions of special types. Our approach is based on the notion of Pareto optimal frontier and on the multi-parameter perturbation the… ▽ More We derive Euler-Lagrange equations for the topology optimization of decay rate in 3-d lossy optical cavities. This leads to a new class of time-harmonic differential or integro-differential equations, which can be written as nonlinear Maxwell systems with switching functions of special types. Our approach is based on the notion of Pareto optimal frontier and on the multi-parameter perturbation theory for eigenfrequencies. Parallels with optimal control theory are discussed. △ Less

Submitted 31 May, 2019; v1 submitted 6 April, 2019; originally announced April 2019.

Comments: 16 pages, the introduction section is extended, the conclusions and discussion section is added, typos are corrected, references are added. OCIS codes: (120.4570) Optical design of instruments; (140.3945) Microcavities; (230.5750) Resonators

MSC Class: 35Q61; 78M50; 49R05 (Primary) 58E17; 90C29; 47A55; 47B44; 47J10; 35B34 (Secondary)

Pareto optimization of resonances and minimum-time control

Authors: Illya M. Karabash, Herbert Koch, Ievgen V. Verbytskyi

Abstract: The aim of the paper is to reduce one spectral optimization problem, which involves the minimization of the decay rate $|\mathrm{Im} \, k |$ of a resonance $k$, to a collection of optimal control problems on the Riemann sphere $\widehat{\mathbb{C}}$. This reduction allows us to apply methods of extremal synthesis to the structural optimization of layered optical cavities. We start from a dual prob… ▽ More The aim of the paper is to reduce one spectral optimization problem, which involves the minimization of the decay rate $|\mathrm{Im} \, k |$ of a resonance $k$, to a collection of optimal control problems on the Riemann sphere $\widehat{\mathbb{C}}$. This reduction allows us to apply methods of extremal synthesis to the structural optimization of layered optical cavities. We start from a dual problem of minimization of the resonator length and give several reformulations of this problem that involve Pareto optimization of the modulus $|k|$ of a resonance, a minimum-time control problem on $\widehat{\mathbb{C}}$, and associated Hamilton-Jacobi-Bellman equations. Various types of controllability properties are studied in connection with the existence of optimizers and with the relationship between the Pareto optimal frontiers of minimal decay and minimal modulus. We give explicit examples of optimal resonances and describe qualitatively properties of the Pareto frontiers near them. A special representation of bang-bang controlled trajectories is combined with the analysis of extremals to obtain various bounds on optimal widths of layers. We propose a new method of computation of optimal symmetric resonators based on minimum-time control and compute with high accuracy several Pareto optimal frontiers and high-Q resonators. △ Less

Submitted 17 December, 2018; v1 submitted 28 August, 2018; originally announced August 2018.

Comments: Section 10 with numerical experiments and Section 11 containing a discussion and the conclusions are added. 5 figures and one table are added. As a consequence,the introduction section is restructured and split into two sections

MSC Class: 49N35; 35B34; 49L25; 78M50; 49R05; 93B27

Journal ref: Journal de Mathématiques Pures et Appliquées 136 (2020), 313-355

Generic asymptotics of resonance counting function for Schrödinger point interactions

Authors: Sergio Albeverio, Illya M. Karabash

Abstract: We study the leading coefficient in the asymptotical formula $ N (R) = \frac{W}π R + O (1) $, $ R \to \infty $, for the resonance counting function $ N (R)$ of Schrödinger Hamiltonians with point interactions. For such Hamiltonians, the Weyl-type and non-Weyl-type asymptotics of $N (R)$ was introduced recently in a paper by J. Lipovský and V. Lotoreichik (2017). In this note, we prove that the Wey… ▽ More We study the leading coefficient in the asymptotical formula $ N (R) = \frac{W}π R + O (1) $, $ R \to \infty $, for the resonance counting function $ N (R)$ of Schrödinger Hamiltonians with point interactions. For such Hamiltonians, the Weyl-type and non-Weyl-type asymptotics of $N (R)$ was introduced recently in a paper by J. Lipovský and V. Lotoreichik (2017). In this note, we prove that the Weyl-type asymptotics is generic. △ Less

Submitted 14 October, 2018; v1 submitted 15 March, 2018; originally announced March 2018.

Comments: 14 pages; the exposition of the proof of Theorem 3.2 is improved, misprints are corrected

MSC Class: 35B34; 35P20 (Primary); 35J10; 05C90; 81Q37; 81Q80 (Secondary)

Journal ref: In: Kurasov P., Laptev A., Naboko S., Simon B. (eds.), "Analysis as a Tool in Mathematical Physics: in Memory of Boris Pavlov'', Oper. Theory Adv. Appl., Vol. 276, pp. 80-93, Birkhäuser, Cham, 2020

Resonance free regions and non-Hermitian spectral optimization for Schrödinger point interactions

Authors: Sergio Albeverio, Illya M. Karabash

Abstract: Resonances of Schrödinger Hamiltonians with point interactions are considered. The main object under the study is the resonance free region under the assumption that the centers, where the point interactions are located, are known and the associated 'strength' parameters are unknown and allowed to bear additional dissipative effects. To this end we consider the boundary of the resonance free regio… ▽ More Resonances of Schrödinger Hamiltonians with point interactions are considered. The main object under the study is the resonance free region under the assumption that the centers, where the point interactions are located, are known and the associated 'strength' parameters are unknown and allowed to bear additional dissipative effects. To this end we consider the boundary of the resonance free region as a Pareto optimal frontier and study the corresponding optimization problem for resonances. It is shown that upper logarithmic bound on resonances can be made uniform with respect to the strength parameters. The necessary conditions on optimality are obtained in terms of first principal minors of the characteristic determinant. We demonstrate the applicability of these optimality conditions on the case of 4 equidistant centers by computing explicitly the resonances of minimal decay for all frequencies. This example shows that a resonance of minimal decay is not necessarily simple, and in some cases it is generated by an infinite family of feasible resonators. △ Less

Submitted 14 August, 2017; v1 submitted 3 August, 2017; originally announced August 2017.

Comments: 23 pages, 1 figure, several typos corrected, acknowledgements updated, these results were reported during the summer school "Modeling, Analysis, and Approximation Theory toward applications in tomography and inverse problems" (Lübeck, July 31 - August 4, 2017)

MSC Class: 35J10; 35B34; 35P15; 49R05; 90C29; 47B44

Journal ref: Operators and Matrices 11 (2017), no.4, 1097-1117

Recovery of periodicities hidden in heavy-tailed noise

Authors: Illya M. Karabash, Jürgen Prestin

Abstract: We address a parametric joint detection-estimation problem for discrete signals of the form $x(t) = \sum_{n=1}^{N} α_n e^{-i λ_n t } + ε_t$, $t \in \mathbb{N}$, with an additive noise represented by independent centered complex random variables $ε_t$. The distributions of $ε_t$ are assumed to be unknown, but satisfying various sets of conditions. We prove that in the case of a heavy-tailed noise i… ▽ More We address a parametric joint detection-estimation problem for discrete signals of the form $x(t) = \sum_{n=1}^{N} α_n e^{-i λ_n t } + ε_t$, $t \in \mathbb{N}$, with an additive noise represented by independent centered complex random variables $ε_t$. The distributions of $ε_t$ are assumed to be unknown, but satisfying various sets of conditions. We prove that in the case of a heavy-tailed noise it is possible to construct asymptotically strongly consistent estimators for the unknown parameters of the signal, i.e., the frequencies $λ_n$, their number $N$, and complex amplitudes $α_n$. For example, one of considered classes of noise is the following: $ε_t$ are independent identically distributed random variables with $\mathbb{E} (ε_t) = 0$ and $\mathbb{E} (|ε_t| \ln |ε_t|) < \infty$. The construction of estimators is based on detection of singularities of anti-derivatives for $Z$-transforms and on a two-level selection procedure for special discretized versions of superlevel sets. The consistency proof relies on the convergence theory for random Fourier series. We discuss also decaying signals and the case of infinite number of frequencies. △ Less

Submitted 20 May, 2017; v1 submitted 29 December, 2015; originally announced December 2015.

Comments: This e-print differs in style, formatting, pagination, and small non-mathematical details from the version of this paper accepted for publication in "Mathematische Nachrichten" (Wiley-VCH). In comparison with the previous version, the following changes have been made: The introduction section 1 was split into two sections 1 and 2. The new section 5 was added. The algorithm 4.1 was added

MSC Class: 94A12; 42A61; 62F12; 42A70; 42A24

Journal ref: Mathematische Nachrichten 291 (2018), no. 1, 86-102

Nonlinear bang-bang eigenproblems and optimization of resonances in layered cavities

Authors: Illya M. Karabash, Olga M. Logachova, Ievgen V. Verbytskyi

Abstract: Quasi-normal-eigenvalue optimization is studied under constraints $b_1(x) \le B(x) \le b_2 (x)$ on structure functions $B$ of 2-side open optical or mechanical resonators. We prove existence of various optimizers and provide an example when different structures generate the same optimal quasi-(normal-)eigenvalue. To show that quasi-eigenvalues locally optimal in various senses are in the spectrum… ▽ More Quasi-normal-eigenvalue optimization is studied under constraints $b_1(x) \le B(x) \le b_2 (x)$ on structure functions $B$ of 2-side open optical or mechanical resonators. We prove existence of various optimizers and provide an example when different structures generate the same optimal quasi-(normal-)eigenvalue. To show that quasi-eigenvalues locally optimal in various senses are in the spectrum $Σ^{nl}$ of the bang-bang eigenproblem $y" = - ω^2 y [ b_1 + (b_2 - b_1) χ_{\mathbb{C}_+} (y^2 ) ]$, where $χ_{\mathbb{C}_+} (\cdot)$ is the indicator function of the upper complex half-plane $\mathbb{C}_+$, we obtain a variational characterization of the nonlinear spectrum $Σ^{nl}$ in terms of quasi-eigenvalue perturbations. To address the minimization of the decay rate $| \mathrm{Im} \ ω|$, we study the bang-bang equation and explain how it excludes an unknown optimal $B$ from the optimization process. Computing one of minimal decay structures for 1-side open settings, we show that it resembles gradually size-modulated 1-D stack cavities introduced recently in Optical Engineering. In 2-side open symmetric settings, our example has an additional centered defect. Nonexistence of global decay rate minimizers is discussed. △ Less

Submitted 19 August, 2015; originally announced August 2015.

Comments: 39 pages, 2 figures, 2 tables. The results of the paper were partially reported at the International Congress of Mathematicians (Seoul ICM 2014) and in several seminar and workshop talks

MSC Class: 49R05; 47J10; 35B34; 49M05; 78M50 (Primary); 46N10; 90C29; 47A55; 70J10; 34L15 (Secondary)

Journal ref: The preprint was published in 2 parts in [Integr. Equ. Oper. Theory 88 (2017), 15-44] and [Methods Funct. Anal. Topology 23 (2017), no.3, 252-260]

Pareto optimal structures producing resonances of minimal decay under $L^1$-type constraints

Authors: Illya M. Karabash

Abstract: Optimization of resonances associated with 1-D wave equations in inhomogeneous media is studied under the constraint $\| B \|_1 <= m$ on the nonnegative function $B \in L^1 (0,\ell)$ that represents the medium's structure. From the Physics and Optimization points of view, it convenient to generalize the problem replacing $B$ by a nonnegative measure $d M$ and imposing on $d M$ the condition that i… ▽ More Optimization of resonances associated with 1-D wave equations in inhomogeneous media is studied under the constraint $\| B \|_1 <= m$ on the nonnegative function $B \in L^1 (0,\ell)$ that represents the medium's structure. From the Physics and Optimization points of view, it convenient to generalize the problem replacing $B$ by a nonnegative measure $d M$ and imposing on $d M$ the condition that its total mass is $<= m$. The problem is to design for a given frequency $α\in R$ a medium that generates a resonance $ ω$ on the line $α+ i R$ with a minimal possible decay rate $| Im ω|$. Such resonances are said to be of minimal decay and form a Pareto frontier. We show that corresponding optimal measures consist of finite number of point masses, and that this result yields non-existence of optimizers for the problem over the set of absolutely continuous measures $B(x) dx$. Then we derive restrictions on optimal point masses and their positions. These restrictions are strong enough to calculate optimal $d M$ if the optimal resonance $ ω$, the first point mass $m_1$, and one more geometric parameter are known. This reduces the original infinitely-dimensional problem to optimization over four real parameters. For small frequencies, we explicitly find the Pareto set and the corresponding optimal measures $d M$. The technique of the paper is based on the two-parameter perturbation method and the notion of local boundary point. The latter is introduced as a generalization of local extrema to vector optimization problems. △ Less

Submitted 11 August, 2013; v1 submitted 8 August, 2013; originally announced August 2013.

Comments: 38 pages, the proof of Lemma 5.4 is corrected, typos are corrected

MSC Class: 49R05; 58E17; 35B34; 34L15; 32A60; 58C06

Journal ref: Journal of Differential Equations 257 (2014) no. 2, 374-414

Nonlinear eigenvalue problem for optimal resonances in optical cavities

Authors: I. M. Karabash

Abstract: The paper is devoted to optimization of resonances in a 1-D open optical cavity. The cavity's structure is represented by its dielectric permittivity function e(s). It is assumed that e(s) takes values in the range 1 <= e_1 <= e(s) <= e_2. The problem is to design, for a given (real) frequency, a cavity having a resonance with the minimal possible decay rate. Restricting ourselves to resonances of… ▽ More The paper is devoted to optimization of resonances in a 1-D open optical cavity. The cavity's structure is represented by its dielectric permittivity function e(s). It is assumed that e(s) takes values in the range 1 <= e_1 <= e(s) <= e_2. The problem is to design, for a given (real) frequency, a cavity having a resonance with the minimal possible decay rate. Restricting ourselves to resonances of a given frequency, we define cavities and resonant modes with locally extremal decay rate, and then study their properties. We show that such locally extremal cavities are 1-D photonic crystals consisting of alternating layers of two materials with extreme allowed dielectric permittivities e_1 and e_2. To find thicknesses of these layers, a nonlinear eigenvalue problem for locally extremal resonant modes is derived. It occurs that coordinates of interface planes between the layers can be expressed via arg-function of corresponding modes. As a result, the question of minimization of the decay rate is reduced to a four-dimensional problem of finding the zeroes of a function of two variables. △ Less

Submitted 25 July, 2012; originally announced July 2012.

Comments: 16 pages

MSC Class: 78M50; 49R05; 47N50; 47A55

Journal ref: Math. Model. Nat. Phenom. 8 (2013), no.1, pp. 143-155

Optimization of quasi-normal eigenvalues for Krein-Nudelman strings

Authors: Illya M. Karabash

Abstract: The paper is devoted to optimization of resonances for Krein strings with total mass and statical moment constraints. The problem is to design for a given $α\in \R$ a string that has a resonance on the line $α+ ı\R$ with a minimal possible modulus of the imaginary part. We find optimal resonances and strings explicitly. The paper is devoted to optimization of resonances for Krein strings with total mass and statical moment constraints. The problem is to design for a given $α\in \R$ a string that has a resonance on the line $α+ ı\R$ with a minimal possible modulus of the imaginary part. We find optimal resonances and strings explicitly. △ Less

Submitted 1 March, 2012; originally announced March 2012.

Comments: 9 pages, these results were extracted in a slightly modified form from the earlier e-print arXiv:1103.4117 [math.SP] following an advise of a journal's referee

MSC Class: 49R05; 78M50; 35P25; 47N50

Journal ref: Integral Equations and Operator Theory 75 (2013), no.2, 235--247

Optimization of quasi-normal eigenvalues for 1-D wave equations in inhomogeneous media; description of optimal structures

Authors: Illya M. Karabash

Abstract: The paper is devoted to optimization of resonances associated with 1-D wave equations in inhomogeneous media. The medium's structure is represented by a nonnegative function B. The problem is to design for a given $α\in \R$ a medium that generates a resonance on the line $α+ ı\R$ with a minimal possible modulus of the imaginary part. We consider an admissible family of mediums that arises in a pro… ▽ More The paper is devoted to optimization of resonances associated with 1-D wave equations in inhomogeneous media. The medium's structure is represented by a nonnegative function B. The problem is to design for a given $α\in \R$ a medium that generates a resonance on the line $α+ ı\R$ with a minimal possible modulus of the imaginary part. We consider an admissible family of mediums that arises in a problem of optimal design for photonic crystals. This admissible family is defined by the constraints $0\leq b_1 \leq B (x) \leq b_2$ with certain constants $b_{1,2}$. The paper gives an accurate definition of optimal structures that ensures their existence. We prove that optimal structures are piecewise constant functions taking only two extreme possible values $b_1$ and $b_2$. This result explains an effect recently observed in numerical experiments. Then we show that intervals of constancy of an optimal structure are tied to the phase of the corresponding resonant mode and write this connection as a nonlinear eigenvalue problem. △ Less

Submitted 24 April, 2012; v1 submitted 21 March, 2011; originally announced March 2011.

Comments: Typos are corrected

MSC Class: 49R05; 49K20; 78M50; 35P25; 47A55; 47N50

Journal ref: Asymptotic Analysis 81 (2013), no. 3-4, pp. 273-295

Bohl-Perron type stability theorems for linear difference equations with infinite delay

Authors: Elena Braverman, Illya M. Karabash

Abstract: Relation between two properties of linear difference equations with infinite delay is investigated: (i) exponential stability, (ii) $ł^p$-input $ł^q$-state stability (sometimes is called Perron's property). The latter means that solutions of the non-homogeneous equation with zero initial data belong to $ł^q$ when non-homogeneous terms are in $ł^p$. It is assumed that at each moment the prehistory… ▽ More Relation between two properties of linear difference equations with infinite delay is investigated: (i) exponential stability, (ii) $ł^p$-input $ł^q$-state stability (sometimes is called Perron's property). The latter means that solutions of the non-homogeneous equation with zero initial data belong to $ł^q$ when non-homogeneous terms are in $ł^p$. It is assumed that at each moment the prehistory (the sequence of preceding states) belongs to some weighted $ł^r$-space with an exponentially fading weight (the phase space). Our main result states that (i) $\Leftrightarrow$ (ii) whenever $(p,q) \neq (1,\infty)$ and a certain boundedness condition on coefficients is fulfilled. This condition is sharp and ensures that, to some extent, exponential and $ł^p$-input $ł^q$-state stabilities does not depend on the choice of a phase space and parameters $p$ and $q$, respectively. $ł^1$-input $ł^\infty$-state stability corresponds to uniform stability. We provide some evidence that similar criteria should not be expected for non-fading memory spaces. △ Less

Submitted 30 September, 2010; originally announced September 2010.

Comments: To be published in Journal of Difference Equations and Applications

MSC Class: 39A11; 39A10; 39A06; 39A12

Journal ref: Journal of Difference Equations and Applications 18 (2012), No. 5, 909-939

A functional model, eigenvalues, and finite singular critical points for indefinite Sturm-Liouville operators

Authors: I. M. Karabash

Abstract: Eigenvalues in the essential spectrum of a weighted Sturm-Liouville operator are studied under the assumption that the weight function has one turning point. An abstract approach to the problem is given via a functional model for indefinite Sturm-Liouville operators. Algebraic multiplicities of eigenvalues are obtained. Also, operators with finite singular critical points are considered. Eigenvalues in the essential spectrum of a weighted Sturm-Liouville operator are studied under the assumption that the weight function has one turning point. An abstract approach to the problem is given via a functional model for indefinite Sturm-Liouville operators. Algebraic multiplicities of eigenvalues are obtained. Also, operators with finite singular critical points are considered. △ Less

Submitted 19 June, 2009; v1 submitted 27 February, 2009; originally announced February 2009.

Comments: 38 pages, Proposition 2.2 and its proof corrected, Remarks 2.5, 3.4, and 3.12 extended, details added in subsections 2.3 and 4.2, section 6 rearranged, typos corrected, references added

MSC Class: 47E05; 34B24; 34B09 (Primary) 34L10; 47B50 (Secondary)

Journal ref: Operator Theory: Advances and Applications, 2010, Volume 203, 247-287

On the nature of ill-posedness of the forward-backward heat equation

Authors: Marina Chugunova, Illya M. Karabash, Sergei G. Pyatkov

Abstract: We study the Cauchy problem with periodic initial data for the forward-backward heat equation defined by the J-self-adjoint linear operator L depending on a small parameter. The problem has been originated from the lubrication approximation of a viscous fluid film on the inner surface of the rotating cylinder. For a certain range of the parameter we rigorously prove the conjecture, based on the… ▽ More We study the Cauchy problem with periodic initial data for the forward-backward heat equation defined by the J-self-adjoint linear operator L depending on a small parameter. The problem has been originated from the lubrication approximation of a viscous fluid film on the inner surface of the rotating cylinder. For a certain range of the parameter we rigorously prove the conjecture, based on the numerical evidence, that the set of eigenvectors of the operator $L$ does not form a Riesz basis in $Ł^2 (-π,π)$. Our method can be applied to a wide range of the evolutional problems given by $PT-$symmetric operators. △ Less

Submitted 27 March, 2008; v1 submitted 17 March, 2008; originally announced March 2008.

Comments: 21 pages; Remark 5.2 added, acknowledgements added, several typos fixed

MSC Class: 35M10; 47B10; 34L40

Journal ref: Integral Equations Operator Theory 65 (2009), no. 3, 319-344

The similarity problem for $J$-nonnegative Sturm-Liouville operators

Authors: Illya M. Karabash, Aleksey S. Kostenko, Mark M. Malamud

Abstract: Sufficient conditions for the similarity of the operator $A := 1/r(x) (-d^2/dx^2 +q(x))$ with an indefinite weight $r(x)=(\sgn x)|r(x)|$ are obtained. These conditions are formulated in terms of Titchmarsh-Weyl $m$-coefficients. Sufficient conditions for the regularity of the critical points 0 and $\infty$ of $J$-nonnegative Sturm-Liouville operators are also obtained. This result is exploited t… ▽ More Sufficient conditions for the similarity of the operator $A := 1/r(x) (-d^2/dx^2 +q(x))$ with an indefinite weight $r(x)=(\sgn x)|r(x)|$ are obtained. These conditions are formulated in terms of Titchmarsh-Weyl $m$-coefficients. Sufficient conditions for the regularity of the critical points 0 and $\infty$ of $J$-nonnegative Sturm-Liouville operators are also obtained. This result is exploited to prove the regularity of 0 for various classes of Sturm-Liouville operators. This implies the similarity of the considered operators to self-adjoint ones. In particular, in the case $r(x)=\sgn x$ and $q\in L^1(R, (1+|x|)dx)$, we prove that $A$ is similar to a self-adjoint operator if and only if $A$ is $J$-nonnegative. The latter condition on $q$ is sharp, i.e., we construct $q\in \cap_{γ<1} L^1(R, (1+|x|)^γdx)$ such that $A$ is $J$-nonnegative with the singular critical point 0. Hence $A$ is not similar to a self-adjoint operator. For periodic and infinite-zone potentials, we show that $J$-positivity is sufficient for the similarity of $A$ to a self-adjoint operator. In the case $q\equiv 0$, we prove the regularity of the critical point 0 for a wide class of weights $r$. This yields new results for "forward-backward" diffusion equations. △ Less

Submitted 19 March, 2008; v1 submitted 10 March, 2008; originally announced March 2008.

Comments: 36 pages, LaTeX2e, version 2; addresses of the authors added, the reference [38] updated

Report number: 1987 (2007), Vienna, Erwin Schroedinger International Institute MSC Class: 47E05; 34B24; 34B09 (Primary) 34L10; 47B50 (Secondary)

Journal ref: J. Differential Equations 246 (2009), no. 3, pp. 964-997

Abstract kinetic equations with positive collision operators

Authors: I. M. Karabash

Abstract: We consider "forward-backward" parabolic equations in the abstract form $Jd ψ/ d x + L ψ= 0$, $ 0< x < τ\leq \infty$, where $J$ and $L$ are operators in a Hilbert space $H$ such that $J=J^*=J^{-1}$, $L=L^* \geq 0$, and $\ker L = 0$. The following theorem is proved: if the operator $B=JL$ is similar to a self-adjoint operator, then associated half-range boundary problems have unique solutions. We… ▽ More We consider "forward-backward" parabolic equations in the abstract form $Jd ψ/ d x + L ψ= 0$, $ 0< x < τ\leq \infty$, where $J$ and $L$ are operators in a Hilbert space $H$ such that $J=J^*=J^{-1}$, $L=L^* \geq 0$, and $\ker L = 0$. The following theorem is proved: if the operator $B=JL$ is similar to a self-adjoint operator, then associated half-range boundary problems have unique solutions. We apply this theorem to corresponding nonhomogeneous equations, to the time-independent Fokker-Plank equation $ μ\frac {\partial ψ}{\partial x} (x,μ) = b(μ) \frac {\partial^2 ψ}{\partial μ^2} (x, μ)$, $ 0<x<τ$, $ μ\in \R$, as well as to other parabolic equations of the "forward-backward" type. The abstract kinetic equation $ T d ψ/dx = - A ψ(x) + f(x)$, where $T=T^*$ is injective and $A$ satisfies a certain positivity assumption, is considered also. △ Less

Submitted 2 October, 2007; v1 submitted 18 August, 2007; originally announced August 2007.

Comments: 20 pages, LaTeX2e, version 2, references have been added, changes in the introduction

MSC Class: 47N55; 35K70; 47B50 (Primary) 35M10; 35K90 (Secondary)

Journal ref: Operator Theory: Advances and Applications, 2009, Volume 188, pp. 175-195

Indefinite Sturm-Liouville operators with the singular critical point zero

Authors: Illya M. Karabash, Aleksey S. Kostenko

Abstract: We present a new necessary condition for similarity of indefinite Sturm-Liouville operators to self-adjoint operators. This condition is formulated in terms of Weyl-Titchmarsh $m$-functions. Also we obtain necessary conditions for regularity of the critical points 0 and $\infty$ of $J$-nonnegative Sturm-Liouville operators. Using this result, we construct several examples of operators with the s… ▽ More We present a new necessary condition for similarity of indefinite Sturm-Liouville operators to self-adjoint operators. This condition is formulated in terms of Weyl-Titchmarsh $m$-functions. Also we obtain necessary conditions for regularity of the critical points 0 and $\infty$ of $J$-nonnegative Sturm-Liouville operators. Using this result, we construct several examples of operators with the singular critical point zero. In particular, it is shown that 0 is a singular critical point of the operator $-\frac{(\sgn x)}{(3|x|+1)^{-4/3}} \frac{d^2}{dx^2}$ acting in the Hilbert space $L^2(\R, (3|x|+1)^{-4/3}dx)$ and therefore this operator is not similar to a self-adjoint one. Also we construct a J-nonnegative Sturm-Liouville operator of type $(\sgn x)(-d^2/dx^2+q(x))$ with the same properties. △ Less

Submitted 6 December, 2006; originally announced December 2006.

Comments: 24 pages, LaTeX2e <2003/12/01>

MSC Class: 47E05; 34B24; 34B09 (Primary) 34L10; 47B50 (Secondary)

Journal ref: Proceedings of the Royal Society of Edinburgh: Section A 138 (2008), no. 4 , pp. 801 -820

Indefinite Sturm-Liouville operators $ (\sgn x) (- \frac{d^2}{dx^2} +q(x))$ with finite-zone potentials

Authors: I. M. Karabash, M. M. Malamud

Abstract: The indefinite Sturm-Liouville operator $A = (\sgn x)(-d^2/dx^2+q(x))$ is studied. It is proved that similarity of $A$ to a selfadjoint operator is equivalent to integral estimates of Cauchy integrals. Also similarity conditions in terms of Weyl functions are given. For operators with a finite-zone potential, the components $\Aess$ and $\Adisc$ of $A$ corresponding to essential and discrete spec… ▽ More The indefinite Sturm-Liouville operator $A = (\sgn x)(-d^2/dx^2+q(x))$ is studied. It is proved that similarity of $A$ to a selfadjoint operator is equivalent to integral estimates of Cauchy integrals. Also similarity conditions in terms of Weyl functions are given. For operators with a finite-zone potential, the components $\Aess$ and $\Adisc$ of $A$ corresponding to essential and discrete spectrums, respectively, are considered. A criterion of similarity of $\Aess$ to a selfadjoint operator is given in terms of Weyl functions for the Sturm-Liouville operator $-d^2/dx^2+q(x)$ with a finite-zone potential $q$. Jordan structure of the operator $\Adisc$ is described. We present an example of the operator $A = (\sgn x)(-d^2/dx^2+q(x))$ such that $A$ is nondefinitizable and $A$ is similar to a normal operator. △ Less

Submitted 13 October, 2006; v1 submitted 2 October, 2006; originally announced October 2006.

Comments: 59 pages, LaTex 2e, Version 3, a mistake in Corollary 5.6 has been corrected, the format of pages has been changed

MSC Class: 47E05; 34B24; 34B09 (Primary) 34L10; 47B50 (Secondary)

Journal ref: Operators and Matrices 1 (2007), no. 3, pp. 301-368