scientific article; zbMATH DE number 2076417
The classical Heisenberg-Weyl uncertainty principle for the Fourier transform on \(\mathbb R\) states that a function and its Fourier transform cannot both be localized with arbitrary precision. The uncertainty principle has been extended to other transforms such as the Hankel transform by \textit{M. Rösler} and \textit{M. Voit} [Proc. Am. Math. Sco. 127, 183--194 (1999; Zbl 0910.44003)] and the Dunkl transform by \textit{M. Rösler} [Bull. Aust. Math. Soc. 59, 353--360 (1999; Zbl 0939.33012)]. In this paper, the authors consider the modified \(Y_\nu\)-transform given by the expression \[ \widetilde f_\nu(x)=\int_0^\infty (xy)^{1-\nu} Y_\nu(xy)f(y)\,dy,\qquad (x\in\mathbb R^+), \] where \(Y_\nu(z)\) is the Bessel function of order \(\nu\) of the second kind, \(f\in L_2(\mathbb R^+,dw_\nu)\) and \(w_\nu\) and \(w_\nu^{-1}\) are measures on \([0,\infty)\) given by \(dw_\nu(x)=x^{1-2\nu}dx\) and \(dw_\nu^{-1}(x)=x^{2\nu-1}dx\). The following uncertainty principle for the modified \(Y_\nu\)-transform is given: \[ \| f(x)\| ^2_{L_2(\mathbb R^+)}\leq C_\nu\| xf(x)\| _{L_2(\mathbb R^+,dw_\nu^{-1})}\| x\widetilde f_\nu(x)\| _{L_2(\mathbb R^+,dw_\nu^{-1})} \] where \(C_\nu\) is a computable constant depends only on \(\nu\), which holds for each \(f\) in a restricted class of the space \(L_2(\mathbb R^+,dw_\nu)\).
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4469874)
