Weighted bounds for the composition of singular integral operators with nonsmooth kernels on spaces of homogeneous type (Q2236650)

From MaRDI portal





scientific article; zbMATH DE number 7413840
Language Label Description Also known as
default for all languages
No label defined
    English
    Weighted bounds for the composition of singular integral operators with nonsmooth kernels on spaces of homogeneous type
    scientific article; zbMATH DE number 7413840

      Statements

      Weighted bounds for the composition of singular integral operators with nonsmooth kernels on spaces of homogeneous type (English)
      0 references
      0 references
      0 references
      25 October 2021
      0 references
      Let \((X,d,\mu)\) be a space of homogeneous type in the sense that \(d\) is a quasi-metric---\(d(x,y)\leq A_0(d(x,z)+d(z,y))\) for fixed \(A_0\geq 1\) -- and there exists \(A_1\) such that for all balls \(B(x,r)\) one has \(\mu(B(x,2r))\leq A_1\mu(B(x,r))\).\par Let \(T\) be bounded on \(L^2(X)\) with kernel \(K\): \(Tf(x)=\int_X K(x,y)\, f(y) d\mu(y)\), \(x\notin {\text{supp}}\, f\). The following assumptions are made.\par Assumption 1. There is an approximation to the identify \(\{A_t\}_{t>0}\) such that \(A_tT\) has an associated kernel \(K_t\) and positive constants \(c_1,c_2,\beta\) such that \[ |K(x,y)-K_t(x,y)|\, \leq \frac{c_2}{\mu(B(x,d(x,y))}\frac{t^{\beta/m}}{\,d(x,y)^\beta},\quad d(x,y)\geq c_1 t^{1/m}\, .\] Assumption 2. There is an approximation to the identify \(\{D_t\}_{t>0}\) such that \(D_tT\) has an associated kernel \(K^t\) and positive constants \(c_3,c_4,\alpha\) such that \[|K^t(x,y)|\leq \frac{c_4}{\mu(B(x,t^{1/m})},\quad d(x,y)\leq c_3 t^{1/m}\, ;\] \[|K^t(x,y)-K(x,y)|\leq \frac{c_4}{\mu(B(x,d(x,y))}\frac{t^{\alpha/m}}{\,d(x,y)^\alpha},\quad d(x,y)\geq c_3 t^{1/m} \, .\] The first main theorem says that if \(T_1\), \(T_2\) are singular integral operators whose kernels satisfy Assumptions 1 and 2, then for \(p\in (1,\infty)\), \(\omega\in A_p\) and \(\sigma=\omega^{-1/p^\prime}\), \[\|T_1 T_2 f\|_{L^p(\omega)}\lesssim [\sigma]_{A_\infty}[\omega]_{A_p}^{1/p}([\sigma]_{A_\infty}+[\omega]_{A_\infty})([\sigma]_{A_p}^{1/p}+[\omega]_{A_\infty}^{1/p^\prime}) \|f\|_{L^p(\omega)}\, .\] The second theorem gives a corresponding weak type bound: under the same assumptions on \(T_1\), \(T_2\), if \(\omega\in A_1\) and \(\lambda>0\) then \[\omega(\{x:|T_1T_2f(x)|>\lambda\})\] \[\lesssim [\omega]_{A_1}[\omega]_{A_\infty}^3 \log^6(e+[\omega]_{A_\infty})\int_X \frac{|f(x)|}{\lambda}\log^2\Bigl(e+ \frac{|f(x)|}{\lambda}\Bigr)\omega(x)\, d\mu(x)\, .\] The constants \([\omega]_{A_p}\) and \([\omega]_{A_\infty}\) are defined in the usual way. The proofs rely on various maximal bounds, including ones for \(M_{L(\log L)^\beta}\) operators, relationships between various quantities that govern doubling measures and \(A_p\) weights, sparse families of dyadic cubes and Carleson families of dyadic cubes, and bilinear and sublinear sparse operators.
      0 references
      Muckenhoupt weight
      0 references
      composition of operators
      0 references
      singular integral
      0 references
      operator with nonsmooth kernel
      0 references
      sparse operator
      0 references
      0 references
      0 references

      Identifiers