Sharp estimates for potential operators associated with Laguerre and Dunkl-Laguerre expansions

DOI10.1007/S11118-015-9501-6MaRDI QIDQ5962760zbMATH OpenOpenAlexWikidataFDO

Publication date 23 February 2016

Published in Potential Analysis (Search for Journal in Brave)

Copyright license Creative Commons Attribution 4.0 International

Full work available at URL https://arxiv.org/abs/1402.2522

Dunkl harmonic oscillator Dunkl-Laguerre expansion fractional integral Laguerre expansion negative power potential kernel potential operator

Mathematics Subject Classification ID

Fractional derivatives and integrals (26A33) Potentials and capacities on other spaces (31C15) Fourier series in special orthogonal functions (Legendre polynomials, Walsh functions, etc.) (42C10) Potential operators (47G40)

Abstract: We study potential operators associated with Laguerre function expansions of convolution and Hermite types, and with Dunkl-Laguerre expansions. We prove qualitatively sharp estimates of the corresponding potential kernels. Then we characterize those

1 l e p, q l e i n f t y

, for which the potential operators are

L^{p} - L^{q}

bounded. These results are sharp analogues of the classical Hardy-Littlewood-Sobolev fractional integration theorem in the Laguerre and Dunkl-Laguerre settings.

Review text

The authors study potential operators in the settings of Laguerre functions expansions of convolution and Hermite types, and Dunkl-Laguerre expansions. The first two situations are associated, respectively, to the second-order differential operators \[ L_{\alpha}=-\frac{d^2}{dx^2}-\frac{2\alpha+1}{x}\frac{d}{dx}+x^2 \] and \[ L_{\alpha}^H=-\frac{d^2}{dx^2}+x^2+\frac{(\alpha-1/2)(\alpha+1/2)}{x^2}\frac{d}{dx}+x^2, \] where \(\alpha>-1\) and \(x>0\). On the other hand, Dunkl-Laguerre expansions are connected to the Dunkl harmonic oscillator in the context of a finite reflection group acting on \(\mathbb{R}\) which is isomorphic to \(\mathbb{Z}_2=\{0,1\}\). The Dunkl harmonic oscillator is a differential-difference operator given by \[ L_{\alpha}^Df(x)=L_{\alpha}f(x)+(\alpha+1/2)\frac{f(x)-f(-x)}{x^2}, \quad \alpha>-1,\,\,x\in \mathbb{R}. \] The parameter \(\alpha\) in the latter is related to the multiplicity function in the Dunkl theory, which turns out to be negative for \(\alpha<-1/2\). This represents an exotic case, due to the importance of the positivity of the multiplicity function in the Dunkl theory. The main results in the paper consist of complete characterizations of the \(L^p-L^q\) boundedness of the potential operators related to the above operators, denoted respectively by \((L_{\alpha})^{-\sigma}\), \((L_{\alpha}^H)^{-\sigma}\) and \((L_{\alpha}^D)^{-\sigma}\), with \(\sigma>0\). Therefore, the authors extent their results of [Can. J. Math. 64, No. 1, 183--216 (2012; Zbl 1244.47043)] to the optimal range of admissible \(p\) and \(q\) in the unweighted one-dimensional case. These results are sharp analogues for the classical Hardy-Littlewood-Sobolev theorem in the context of Laguerre and Dunkl-Laguerre settings. The proofs rely on qualitatively sharp estimates for the corresponding potential kernels \(K^{\alpha,\sigma}\), \(K_H^{\alpha,\sigma}\) and \(K_D^{\alpha,\sigma}\). The Hermite type kernel \(K_H^{\alpha,\sigma}\) and the Dunkl-Laguerre kernel \(K_D^{\alpha,\sigma}\) are shown to be related to the convolution type kernel \(K^{\alpha,\sigma}\). In this way, the first main effort is devoted to proving sharp estimates for \(K^{\alpha,\sigma}\). This work is carried out by means of a careful study of the integral expression representing \(K^{\alpha,\sigma}\), which involves modified Bessel functions and hyperbolic functions. Then, the \(L^p-L^q\) boundedness result for \((L_{\alpha})^{-\sigma}\) is proved, by an appropriate splitting of the operator and the use of the sharp estimates for the kernel. Other additional ingredients are some known results for the potential operator associated to a Jacobi differential operator, and a duality argument based on the symmetry of the kernel and the Riesz-Thorin interpolation theorem. The study of suitable counterexamples leads to negative results, and this yields the sharpness of the result. The proof of \(L^p-L^q\) estimates for \((L_{\alpha}^H)^{-\sigma}\) follows the ideas of the proof for \((L_{\alpha})^{-\sigma}\), with suitable adaptations, and for \((L_{\alpha}^D)^{-\sigma}\) the relation between kernels is used to get the mapping properties in a direct way. Finally, it is remarkable that qualitatively sharp estimates for \(K_D^{\alpha,\sigma}\) are also provided when \(\alpha\geq-1/2\), although they are not needed for the corresponding \(L^p-L^q\) result. When \(\alpha<-1/2\), the kernel attains negative values, so general sharp estimates are not possible.

Recommendations

Cites work

Cited in

(7)

This page was built for publication: Sharp estimates for potential operators associated with Laguerre and Dunkl-Laguerre expansions

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5962760)