0
$\begingroup$

In the appendix of the paper (https://arxiv.org/abs/math/0604185) the authors said "The Riesz transform are singular integral operators with kernels $K(r,\zeta)=r^{-2}\Omega(\zeta)$, where $(r,\zeta)$ are the polar coordinates. The function $\Omega$ is smooth and $\int_{S^1} \Omega(\zeta)d\sigma(\zeta)=0$." Can anyone explain a little bit more about this statement. I don't quite understand how this is related to the definition of Riesz transform by using Fourier multiplier method. Here the authors consider in $\mathbb R^2$.

$\endgroup$
1
$\begingroup$

We define the Riesz transform, say, in the first coordinate by \begin{align*} R_{1}(f)(x)=(f\ast W_{1})(x), \end{align*} where \begin{align*} \left<W_{1},\varphi\right>=\lim_{\epsilon\rightarrow 0}\int_{|y|\geq\epsilon}\dfrac{y_{1}}{|y|^{3}}\varphi(y)dy, \end{align*} so \begin{align*} (f\ast W_{1})(x)=\left<W_{1},f(x-\cdot)\right>=\lim_{\epsilon\rightarrow 0}\int_{|x-y|\geq\epsilon}\dfrac{x_{1}-y_{1}}{|x-y|^{3}}f(y)dy. \end{align*} However, the expression can be written as \begin{align*} (f\ast W_{1})(x)=\lim_{\epsilon\rightarrow 0}\int_{|x-y|\geq\epsilon}\dfrac{\Omega((x-y)/|x-y|)}{|x-y|^{2}}f(y)dy, \end{align*} where $\Omega$ is defined on the sphere ${\bf{S}}^{1}$, $\Omega(\theta)=\theta_{1}$, we now let $K(r,\theta)=r^{-2}\Omega(\theta)$, then \begin{align*} (f\ast W_{1})(x)=\lim_{\epsilon\rightarrow 0}\int_{{\bf{S}}^{1}}\int_{\epsilon}^{\infty}\dfrac{\Omega(\theta)}{r^{2}}f(x-r\theta)rdrd\theta=\lim_{\epsilon\rightarrow 0}\int_{{\bf{S}}^{1}}\int_{\epsilon}^{\infty}K(r,\theta)f(x-r\theta)rdrd\theta. \end{align*} On the other hand, we have \begin{align*} \int_{{\bf{S}}^{1}}\Omega(\theta)d\theta=\int_{{\bf{S}}^{1}}\theta_{1}d\theta=0 \end{align*} since $\theta\rightarrow\theta_{1}$ is odd.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.