Separation of Eigenvalues of the Wave Equation for the Unit Ball in RN

It is shown that π is the infinium gap between the consecutive square roots of the eigenvalues of the wave equation in a hypespherical domain for both the Neumann (free) and the full range of mixed (elastic) homogeneous boundary conditions. Previous literature contains the same information apparently only for the Dirichlet (fixed) boundary condition. These square roots of the eigenvalues are the zeros of solutions of a differential equation in Bessel functions (first kind) and their first derivatives. The infinium gap is uniform for Bessel functions of orders x ≥ ½ (as well as for x = 0). The intervals between the roots are actually monotone decreasing in length. These results are obtained by interlacing zeros of Bessel and associated functions and comparing their relative displacements with oscillation theory. If W_l denotes the lth positive root for some fixed order x, the minimum gap property assures that {exp(±iw_lt|l = 1, 2,...} form a Riesz basis in L²(0, τ) for τ > 2. This has application to the problem of controlling solutions of the wave equation by controlling the boundary values.     

On a function associated with the Bessel functions

Summary Defining the function Δn, 1,k;x(J) asΔn, 1,k;x(J)=J _n+1(x)−J _n(x)J _n+k+1(x) associated with the Bessel functionJ _n(x), we derive a series of products of Bessel functions for Δ_{n, f, k, x} (J). Whenk=1,k;x (J) becomes Turàn expression for Bessel functions. Some consequences have been pointed out.

---

Riassunto Definita la Δ_{n, f, k, x} (J) come Δ_{n, f, k, x}, (J)=J _n+1(x)J _n+k(x)-J _n(_n+k+1)(x) associata alla funzioneJ _n(x) di Bessel, si ricava una serie di prodotti di funzioni di Bessel per Δ_{n, f, k, x}, (J). 3 Quandok=1, Δ_{n, f, k, x}, (J) diventa una espressione di Turàn per le funzioni di 2 Bessel, vengono inoltre indicate alcune altre conseguenze.

     

La riflessione analitica delle funzioni biarmoniche attorno a un cerchio ed alcuni problemi di elasticità piana

Sunto Due funzionif(x, y) e ϕ(x, y), biarmoniche (e cioè soddisfacenti all'equazione ΔΔ=0), rispettivamente definite nei semipianix>0 edx<0, le cui derivate seconde si annullano all'infinito, e tali che nei punti dell'assey risultif=ϕ e∂f/∂x=∂ϕ/∂x, si dicono l'una ? riflessa ? dell'altra attorno all'assey. Da ognuna delle due funzionif e ϕ l'altra si ottiene con sole operazioni di sostituzione e derivazione (indicando, precisamente, con{f}, {∂f/∂x} e{Δf} le funzioni che si ottengono daf, ∂f/∂x eΔf ponendo, in queste, in luogo dix il suo contrario −x, si ha ϕ={f}+2x{∂f/∂x}+x ² {Δf} e, reciprocamente,f={ϕ}+2x{∂ϕ/∂x}+x ²{Δϕ}). In modo analogo si definisce una operazione di riflessione analitica attorno a un cerchio. La retta potendosi riguardare come cerchio degenere (di raggio infinito) l'operazione di rifiessione analitica attorno alla retta viene ottenuta, nel testo, come caso limite di quella di riflessione attorno al cerchio. L'operazione di riflessione analitica trova applicazione in alcuni problemi di elasticità piana (perturbazione prodotta da un foro circolare nella sollecitazione di un sistema piano; determinazione degli sforzi in un semipiano elastico sollecitato da una forza applicata in un punto interno).     

A generalization of an example by V. Ya. Ivrii

A well-known example by Ivrii concerning the operatorP=D _t ² −t^2lD _x ² +at^kD_x (a≠0), shows that there exists a delicate relation amongl, k and the Gevrey index of well-posedness of the Cauchy problem. In this paper we give a generalization to a class of pseudo-differential operators includingP.

---

Sunto Un famoso esempio di Ivrii riguardante l'operatoreP=D _t ² −t^2lD _x ² +at^kD_x (a≠0), mostra che c'è una relazione sottile tral, k e l'indice di Gevrey di buona positura del problema di Cauchy. In questo articolo viene data una generalizzazione ad una classe di operatori pseudodifferenziali che comprendeP.

     

Inertial manifold for a hyperbolic equation with dissipation

Sufficient conditions for the existence of an inertial manifold are found for the equation u _tt + 2γu _t − Δu = f(u, u _t ), u = u(x, t), x ∈ Ω ⋐ ℝ ^N , u| _∂Ω = 0, t > 0 under the assumption that the function f satisfies the Lipschitz condition.     

Extinction of solutions for some nonlinear parabolic equations

We are dealing with the first vanishing time for solutions of the Cauchy–Neumann problem for the semilinear parabolic equation ∂ _t u − Δu + a(x)u ^q = 0, where a(x) \geqslant d₀exp( - \fracw( | x | )| x |² ) a(x) \geqslant {d_0}\exp \left( { - \frac{{\omega \left( {\left| x \right|} \right)}}{{{{\left| x \right|}^2}}}} \right) , d ₀ > 0, 1 > q > 0, and ω is a positive continuous radial function. We give a Dini-like condition on the function ω which implies that any solution of the above equation vanishes in finite time. The proof is derived from semi-classical limits of some Schr¨odinger operators.     

Characteristic features of the dynamics of the Ginzburg-Landau equation in a plane domain

We study the boundary value problem w_t=ℵ₀Δw+ℵ₁w-ℵ₂w|w|²,w|∂Ω₀=0 in the domain Ω₀={(x,y):0 ≤ x ≤ l₁,0 ≤ y ≤ l₂}. Here, w is a complex-valued function, Δ is the laplace operator, and ℵ_j, j=0,1,2, are complex constants withRe ℵ_j > 0. We show that under a rather general choice of the parameters l₁ and l₂, the number of stable invariant tori in the problem, as well as their dimensions, grows infinitely asRe ℵ₀ → 0 andRe ℵ₀ → 0. Translated from Teoreticheskaya i Matematicheskaya Fizika, Vol. 125, No. 2, pp. 205–220, November, 2000.     

Pseudodifferential Operators on Local Hardy Spaces

In this work, we study the continuity of pseudodifferential operators on local Hardy spaces h ^p (ℝ ⁿ ) and generalize the results due to Goldberg and Taylor by showing that operators with symbols in S _1,δ ⁰(ℝ ⁿ ), 0≤δ<1, and in some subclasses of S _1,1⁰(ℝ ⁿ ) are bounded on h ^p (ℝ ⁿ ) (0<p≤1). As an application, we study the local solvability of the planar vector field L=∂ _t +ib(x,t)∂ _x , b(x,t)≥0, in spaces of mixed norm involving Hardy spaces. Work supported in part by CNPq, FINEP, and FAPESP.     

Asymptotic behaviour of a Brownian motion on exterior domains

We study the asymptotic behaviour of the transition density of a Brownian motion in ?, killed at ∂?, where ? ^c is a compact non polar set. Our main result concern dimension d = 2, where we show that the transition density p ^? _t (x, y) behaves, for large t, as u(x)u(y)(t(log t)²)⁻¹ for x, y∈?, where u is the unique positive harmonic function vanishing on (∂?) ^r , such that u(x) ∼ log ∣x∣. Received: 29 January 1999 / Revised version: 11 May 1999     

Correlation structure of intermittency in the parabolic Anderson model

Consider the Cauchy problem ∂u(x, t)/∂t = ℋu(x, t) (x∈ℤ^d, t≥ 0) with initial condition u(x, 0) ≡ 1 and with ℋ the Anderson Hamiltonian ℋ = κΔ + ξ. Here Δ is the discrete Laplacian, κ∈ (0, ∞) is a diffusion constant, and ξ = {ξ(x): x∈ℤ ^d } is an i.i.d.random field taking values in ℝ. G?rtner and Molchanov (1990) have shown that if the law of ξ(0) is nondegenerate, then the solution u is asymptotically intermittent. In the present paper we study the structure of the intermittent peaks for the special case where the law of ξ(0) is (in the vicinity of) the double exponential Prob(ξ(0) > s) = exp[−e ^{s /θ}] (s∈ℝ). Here θ∈ (0, ∞) is a parameter that can be thought of as measuring the degree of disorder in the ξ-field. Our main result is that, for fixed x, y∈ℤ ^d and t→∈, the correlation coefficient of u(x, t) and u(y, t) converges to ∥w _ρ∥⁻² _ℓ2Σ_{z ∈ℤd} w _ρ(x+z)w _ρ(y+z). In this expression, ρ = θ/κ while w _ρ:ℤ^d→ℝ⁺ is given by w _ρ = (v _ρ)^{⊗ d} with v _ρ: ℤ→ℝ⁺ the unique centered ground state (i.e., the solution in ℓ²(ℤ) with minimal l ²-norm) of the 1-dimensional nonlinear equation Δv + 2ρv log v = 0. The uniqueness of the ground state is actually proved only for large ρ, but is conjectured to hold for any ρ∈ (0, ∞). empty It turns out that if the right tail of the law of ξ(0) is thicker (or thinner) than the double exponential, then the correlation coefficient of u(x, t) and u(y, t) converges to δ _{x, y} (resp.the constant function 1). Thus, the double exponential family is the critical class exhibiting a nondegenerate correlation structure. Received: 5 March 1997 / Revised version: 21 September 1998     

Weakly Singular Integral Operators in Weighted L∞–Spaces

We study integral operators on (−1, 1) with kernels k(x, t) which may have weak singularities in (x, t) with x ∈N₁, t ∈N₂, or x=t, where N₁,N₂ are sets of measure zero. It is shown that such operators map weighted L^∞–spaces into certain weighted spaces of smooth functions, where the degree of smoothness is the higher the smoother the kernel k(x, t) as a function in x is. The spaces of smooth function are generalizations of the Ditzian-Totik spaces which are defined in terms of the errors of best weighted uniform approximation by algebraic polynomials.     

A variational approach in weighted Sobolev spaces to the operator −Δ+∂/∂x 1 in exterior domains of ℝ3

Consider the Dirichlet problem −vΔu+k∂ ₁ u = f withv, k>0 in ℝ³ or in an exterior domain of ℝ³ where the skew-symmetric differential operator −₁=∂/∂x₁ is a singular perturbation of the Laplacian. Because of the inhomogeneity of the fundamental solution we study existence, uniqueness and regularity in Sobolev spaces with anisotropic weights. In these spaces the operator ∂₁ yields an additional positive definite term giving better results than in Sobolev spaces with radial weights. The elliptic equation −vΔu +k∂₁ u=f can be taken as a model problem for the Oseen equations, a linearized form of the Navier-Stokes equations. Supported by the Sonderforschungsbereich 256 of the Deutsche Forschungsgemeinschaft at the University of Bonn     

Zero-Eigenvalue Eigenfunctions for Differences of Elliptic Relativistic Calogero-Moser Hamiltonians

Letting A_l(x) denote the commuting analytic difference operators of elliptic relativistic Calogero-Moser type, we present and study zero-eigenvalue eigenfunctions for the operators A_l(x) − A_l(−y) (with l = 1, 2,..., N and x, y ∈ ℂ ^N). The eigenfunctions are products of elliptic gamma functions. They are invariant under permutations of x₁,..., x_N and y₁,..., y_N and under interchange of the step-size parameters. __________ Translated from Teoreticheskaya i Matematicheskaya Fizika, Vol. 146, No. 1, pp. 31–41, January, 2006.     

Functional equations for Hecke-Maaass series

The Dirichlet (Hecke-Maass) series associated with the eigenfuctionsf andg of the invariant differential operator Δ_k=−y²(∂²/∂x²)+iky∂/∂x of weightk are investigated. It is proved that any relation of the form (f/kM)=g for thek-action of the groupSL ₂ SL ₂(ℝ) is equivalent to a pair of functional equations relating the Hecke-Maass series forf andg and involving only traditional gamma factors. This work was supported by the Russian Foundation for Basic Research (grant No. 96-01-10439). Institute of Applied Mathematics, Far East Division of Russian Academy of Sciences. Translated from Funktional'nyi Analiz i Ego Prilozheniya, Vol. 34, No. 2, pp. 23–32, April–June, 2000. Translated by V. M. Volosov     

Multiscale stochastic homogenization of convection-diffusion equations

Multiscale stochastic homogenization is studied for convection-diffusion problems. More specifically, we consider the asymptotic behaviour of a sequence of realizations of the form ∂u _ɛ ^ω / ∂t+1 / ɛ ₃ C(T ₃(x/ɛ ₃)ω ₃) · ∇u _ɛ ^ω − div(α(T ₂(x/ɛ ₂)ω ₂, t) ∇u _ɛ ^ω ) = f. It is shown, under certain structure assumptions on the random vector field C(ω ₃) and the random map α(ω ₁, ω ₂, t), that the sequence {u _ɛ ^ω } of solutions converges in the sense of G-convergence of parabolic operators to the solution u of the homogenized problem ∂u/∂t − div (B(t)∇u= f).     

Diffusion approximation for the convection-diffusion equation with random drift

We consider the asymptotic behavior of the solutions ofscaled convection-diffusion equations ∂ _t u _ɛ (t, x) = κΔ _x (t, x) + 1/ɛV(t/ɛ²,xɛ) ·∇ _x u _ɛ (t, x) with the initial condition u _ɛ(0,x) = u ₀(x) as the parameter ɛ↓ 0. Under the assumptions that κ > 0 and V(t, x), (t, x) ∈R ^d is a d-dimensional,stationary, zero mean, incompressible, Gaussian random field, Markovian and mixing in t we show that the laws of u _ɛ(t,·), t≥ 0 in an appropriate functional space converge weakly, as ɛ↓ 0, to a δ-type measureconcentrated on a solution of a certain constant coefficient heat equation. Received: 23 March 2000 / Revised version: 5 March 2001 / Published online: 9 October 2001     

The Upper Bounds of Arbitrary Eigenvalues for Uniformly Elliptic Operators with Higher Orders

Let Ωbelong to R^m （m≥ 2） be a bounded domain with piecewise smooth and Lipschitz boundary δΩ Let t and r be two nonnegative integers with t ≥ r ＋ 1. In this paper, we consider the variable-coefficient eigenvalue problems with uniformly elliptic differential operators on the left-hand side and （-Δ）^T on the right-hand side. Some upper bounds of the arbitrary eigenvalue are obtained, and several known results are generalized.     

Almost sure asymptotics for the continuous parabolic Anderson model

We consider the parabolic Anderson problem ∂ _t u = κΔu + ξ(x)u on ℝ₊×ℝ ^d with initial condition u(0,x) = 1. Here κ > 0 is a diffusion constant and ξ is a random homogeneous potential. We concentrate on the two important cases of a Gaussian potential and a shot noise Poisson potential. Under some mild regularity assumptions, we derive the second-order term of the almost sure asymptotics of u(t, 0) as t→∞. Received: 26 July 1999 / Revised version: 6 April 2000 / Published online: 22 November 2000     

Euler operator and homogeneous hida distributions

Let (S)⊄L ²(S′(∔),μ)⊄(S)^* be the Gel'fand triple over the white noise space (S′(∔),μ). Let (e _n ,n>-0) be the ONB ofL ²(∔) consisting of the eigenfunctions of the s.a. operator . In this paper the Euler operator Δ _E is defined as the sum , where ∂ _i stands for the differential operatorD _{e i}. It is shown that Δ _E is the infinitesimal generator of the semigroup (T _t ), where (T _t ϕ)(x)=ϕ(e ^t x) for ϕ∈(S). Similarly to the finite dimensional case, the λ-order homogeneous test functionals are characterized by the Euler equation: Δ _Eϕ =λ_ϕ. Via this characterization the λ-order homogeneous Hida distributions are defined and their properties are worked out. Supported by the National Natural Science Foundation of China.     

Reduced measures associated with parabolic problems

We study the existence and the properties of reduced measures for the parabolic equations ∂ _t u − Δu + g(u) = 0 in Ω × (0, ∞) subject to the conditions (P): u = 0 on ∂Ω × (0, ∞), u(x, 0) = μ and (P′): u = μ′ on ∂Ω × (0, ∞), u(x, 0) = 0, where μ and μ′ are positive Radon measures and g is a continuous nondecreasing function.