Universal \(L^p\) improving for averages along polynomial curves in low dimensions (Q837060)

From MaRDI portal





scientific article
Language Label Description Also known as
default for all languages
No label defined
    English
    Universal \(L^p\) improving for averages along polynomial curves in low dimensions
    scientific article

      Statements

      Universal \(L^p\) improving for averages along polynomial curves in low dimensions (English)
      0 references
      0 references
      0 references
      0 references
      10 September 2009
      0 references
      Near sharp \(L^p\to L^q\) boundedness of averaging operators along polynomial curves is proved in two and three dimensions. The results provide an alternate proof and extension of results of \textit{D. M. Oberlin} [Math. Scand. 90, No. 1, 126--138 (2002; Zbl 1034.42012)]. Given a smooth curve \(\gamma:I\to \mathbb{R}^d\) one defines the affine arclength measure \(\nu=\nu_\gamma \) by its action on test functions as \[ \nu(\phi) = \int_I \phi(\gamma(t)) |L_\gamma (t)|^{2\over d(d+1)}\, dt \] where \(L_\gamma (t)=\text{det}\, (\gamma'(t),\dots, \gamma^{(d)}(t))\). The measure \(\nu\) is invariant under reparametrizations of \(\gamma\). One defines the convolution along \(\gamma\) by \[ T_\gamma f(x) = f\ast\nu (c) =\int_I f(x-\gamma(t)) |L_\gamma (t)|^{2\over d(d+1)}\, dt. \] Here, the problem of obtaining bounds for polynomial curves \({\mathbf{P}(t)}=(P_1 (t),\dots, P_d(t))\) with bounds depending only on the degree of the polynomials in dimensions two and three is considered. Actually, different arguments are needed \(d=2\) and \(d=3\). The main theorem states that for \(d=2,3\) and for any \(\epsilon>0\) there exists a \(C\) that depends only on \(\epsilon\), the dimension, and the degree of \(\mathbf{P}\) but not on \(I\) such that \[ \|\mathcal{A}\|_{L^{{d^2+d\over 2d-2},{d+1\over 2}+\epsilon}}\leq C\|f\|_{L^{d+2\over 2}} \] and \[ \|\mathcal{A}\|_{L^{{d+1\over d-1},{d^2+d\over d^2-d+2}+\epsilon}}\leq C\|f\|_{L^{d^2+d\over d^2-d+2}}. \] The norms on the left are Lorentz space norms. The fact that the result cannot be pushed to \(\epsilon=0\) follows from general homogeneity properties. The proofs combine a combinatorial method due to \textit{M. Christ} [Int. Math. Res. Not. 1998, No. 19, 1033--1048 (1998; Zbl 0927.42008)] together with a geometric inequality for vector polynomials due to the authors in earlier work [C. R., Math., Acad. Sci. Paris 346, No. 1--2, 45--48 (2008; Zbl 1175.42008)].
      0 references
      averaging operators
      0 references
      polynomial curves
      0 references
      universal bounds
      0 references
      0 references
      0 references
      0 references

      Identifiers