Blocking analysis and \(T(1)\) theorem (Q1267312)

Let \(T\) be a linear continuous mapping from \({\mathcal D}(\mathbb{R}^n)\) to \({\mathcal D}'(\mathbb{R}^n)\) associated with a kernel \(K(x,y)\). The celebrated \(T(1)\) theorem by David and Journé tells us that if the kernel \(K\) satisfies certain size and smoothness conditions then \(T\) extends a bounded operator on \(L^2(\mathbb{R}^n)\) if and only if \(T(1)\in\text{BMO}\), \(T^*(1)\in\text{BMO}\) and \(T\) satisfies the weak boundedness property. This theorem was improved by Meyer, who showed that the size and smoothness conditions on \(K\) can be weakened by the following integral estimates: \[ \sup_{r>0} \int_{r\leq| x-y|< 2r} (| K(x,y)|+| K(y,x)|) dy\leq C, \] \[ \sup_{\substack{ r>0\\ | u|+| v|\leq r}} \Biggl\{\int_{2^kr\leq| x-y|< 2^{k+1}r}| K(x+ u,y+v)- K(x,y)| (dx+ dy)\Biggr\}= B(k), \] where \(k= 1,2,\dots\) and \(B(k)\) satisfies \(\sum k B(k)< \infty\). In this paper, the authors improved Meyer's result by using a weaker condition \(\sum k^{1/2}B(k)< \infty\) instead of \(\sum k B(k)< \infty\) to obtain the same \(L^2\) boundedness theorem. It is worthy pointing out that the proof in this paper is based on the theory of wavelet analysis and it is different from Meyer's proof.