An alternate approach to bilinear rough singular integrals
Ankit Bhojak, Saurabh Shrivastava · 2025
Read An alternate approach to bilinear rough singular integrals by Ankit Bhojak, Saurabh Shrivastava (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2508-19181.txt.
Introduction
Ankit Bhojak and Saurabh Shrivastava take a fresh path to rough two-input singular sums in a 2025 paper. These sums are a hard case in modern analysis, and past proofs leaned on wave-split tools. The new path expands the inputs in local wave series and lands on three-input bounds that fade at the right rate.
With it, the pair prove sharp bounds for the cut-off form of the sums when the core sits away from the match line. The same path yields a self-standing proof of bounds in all dimensions for the best range of powers. The method may ease work on kin problems too.
Catalog record
| CREATOR | Ankit Bhojak, Saurabh Shrivastava |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 87 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2508-19181
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
An alternate approach to bilinear rough singular integrals Ankit Bhojak, Saurabh Shrivastava Abstract . The goal of this paper is to provide a new approach to address the L p − L^{p}- boundedness of bilinear rough singular integral operators. This approach relies on local Fourier series expansion of input functions leading to trilinear estimates with desired decay in the frequency parameter. This approach departs from the existing methods of the wavelet decomposition of the multiplier employed in the work of Grafakos, He and Honzík and in a series of subsequent papers in the context of bilinear rough singular integrals. With this new approach, we provide a new and self contained proof of L p − L^{p}- boundedness of bilinear rough singular integral operators in dimension one for the optimal range of exponents. Furthermore, this approach allows us to prove sharp L p − L^{p}- estimates for maximally truncated bilinear rough singular integrals when the kernel is supported away from the diagonal in the plane. 2010 Mathematics Subject Classification Primary 42B20, 42B25 1. Introduction The study of singular integrals of the form Tf(x)=p.v.\int K(y)f(x-y)dy is one of the central themes in harmonic analysis. For the kernel K(y)=\Omega(\frac{y}{|y|})|y|^{-d} with \Omega\in L\log L({\mathbb{S}}^{d-1}) , satisfying mean zero condition, the boundedness of the corresponding operator T on L^{p},\;1<p<\infty dates back to the work of Calderón and Zygmund [ 2 ] . A different proof of the L^{p}- estimates of Tf , based on a double dyadic decomposition of the kernel and the Fourier transform estimates, was obtained by Duoandikoetxea and Rubio de Francia [ 10 ] . The issue of weak type estimates at the endpoint p=1 was addressed by Christ [ 4 ] and Hoffman [ 22 ] separately for dimension two and later by Seeger [ 32 ] for all dimensions. Let \Omega\in L^{1}({\mathbb{S}}^{2d-1}) be with mean value zero condition \int_{{\mathbb{S}}^{2d-1}}\Omega(\theta)d\sigma(\theta)=0 , where d\sigma is the surface measure on {\mathbb{S}}^{2d-1} . The bilinear rough singular integral operator T_{\Omega} is defined as T_{\Omega}(f_{1},f_{2})(x)=\int_{{\mathbb{R}}^{2d}}\Omega\left(\frac{(y_{1},y_{2})}{|(y_{1},y_{2})|}\right){|(y_{1},y_{2})|^{-2d}}f_{1}(x-y_{1})f_{2}(x-y_{2})\;dy_{1}dy_{2}. Coifman and Meyer [ 6 , Theorem I] showed that if \Omega is a function of bounded variation on {\mathbb{S}}^{1} , then T_{\Omega} is bounded from L^{p_{1}}({\mathbb{R}})\times L^{p_{2}}({\mathbb{R}}) to L^{p}({\mathbb{R}}) for (p_{1},p_{2},p)\in\mathcal{H}^{1} , where \mathcal{H}^{q}=\left\{(p_{1},p_{2},p):\;1<p_{1},p_{2}<\infty,\;\frac{1}{2}<p<\infty,\;\frac{1}{p_{1}}+\frac{1}{p_{2}}=\frac{1}{p},\;\text{and}\;\frac{1}{p}+\frac{1}{q}<2\right\},\;1\leq q\leq\infty. In higher dimensions, Kenig and Stein [ 25 ] and Grafakos and Torres [ 18 ] proved the boundedness of T_{\Omega} , for \Omega being a Lipschitz function, in the larger range of exponents in \mathcal{H}^{\infty} . We note that when \Omega is an odd function on {\mathbb{S}}^{1} , the operator T_{\Omega} can be expressed as T_{\Omega}(f_{1},f_{2})(x)=\int_{{\mathbb{S}}^{1}}\Omega(\theta_{1},\theta_{2})BH_{\theta_{1},\theta_{2}}(f_{1},f_{2})(x)d\sigma(\theta_{1},\theta_{2}), where BH_{\theta_{1},\theta_{2}} is the bilinear Hilbert transform defined by BH_{\theta_{1},\theta_{2}}(f_{1},f_{2})(x)=\int_{\mathbb{R}}f_{1}(x-\theta_{1}t)f_{2}(x-\theta_{2}t)\frac{dt}{t}. The boundedness of BH_{\theta_{1},\theta_{2}} was proved in the seminal works of Lacey and Thiele [ 26 , 27 ] . By exploiting the uniform boundedness of H_{\theta_{1},\theta_{2}} [ 17 , 28 ] , the authors in [ 7 ] established the boundedness of T_{\Omega} for (p_{1},p_{2},p)\in\mathcal{H}^{1} satisfying the condition \max\left\{\left|\frac{1}{p_{1}}-\frac{1}{p_{2}}\right|,\left|\frac{1}{p_{2}}-\frac{1}{p^{\prime}}\right|,\left|\frac{1}{p^{\prime}}-\frac{1}{p_{1}}\right|\right\}<\frac{1}{2} whenever the even part of \Omega lies in the Hardy space H^{1}({\mathbb{S}}^{1}) . The first complete result for the bilinear operator T_{\Omega} without any smoothness assumption on \Omega was obtained by Grafakos, He and Honzík in [ 14 ] . They showed that T_{\Omega} is bounded for exponents in the range \mathcal{H}^{\infty} for \Omega\in L^{\infty}({\mathbb{S}}^{2d-1}) . Their breakthrough idea was to apply a double dyadic decomposition of the kernel as in [ 10 ] and to decompose the multiplier further by employing a tensor-type compactly supported wavelet basis with vanishing moments. These ideas coupled with combinatorial arguments led them to an L^{2}\times L^{2}\to L^{1}- estimate with appropriate decay in the frequency parameter whenever \Omega\in L^{2}({\mathbb{S}}^{2d-1}) . Further, since the bilinear Calderón-Zygmund theory from [ 18 ] behaves well for \Omega\in L^{\infty} , an interpolation argument for bilinear operators applied to the two type of estimates mentioned as above, leads to the sharp L^{p}- boundedness results for T_{\Omega} when \Omega\in L^{\infty} . Grafakos, He and Honzík [ 14 ] also proved the L^{p}- boundedness of T_{\Omega} in the local L^{2} range: 2<p_{1},p_{2},<\infty,\;1\leq p\leq 2 for \Omega\in L^{2} . This result was further extended to all \Omega\in L^{q} for q>\frac{4}{3} in [ 16 ] . Building upon the (2,2,1)- inequality from [ 14 ] , He and Park [ 21 ] improved the range of boundedness of T_{\Omega} to exponents in \mathcal{H}^{q} for \Omega\in L^{q},\;q>\frac{4}{3} . In doing so, they developed bilinear Calderón-Zygmund theory further which is suitable for all 1<q\leq\infty . We refer the reader to [ 11 , 12 ] for multilinear extensions of this result. In [ 15 ] , the authors showed that \mathcal{H}^{q} is the largest open range for which the operator T_{\Omega} is bounded for \Omega\in L^{q},\;1<q\leq\infty . Finally, Dosidis and Slavíková [ 9 ] showed that T_{\Omega} is also bounded in the range \mathcal{H}^{q} whenever \Omega\in L^{q},\;1<q\leq\infty . In this work, they utilized the boundedness of shifted square functions, whose details may be found in [ 29 ] , to obtain an admissible growth estimates for Littlewood-Paley pieces of the operator T_{\Omega} in the Banach range of exponents for \Omega\in L^{1} . They also showed that the strong type boundedness of T_{\Omega} cannot hold outside the region \mathcal{H}^{q} for 1\leq p_{1},p_{2}<\infty . However, the weak type boundedness of T_{\Omega} on the lines p_{1}=1 , p_{2}=1 and \frac{1}{p}+\frac{1}{q}=2 remains open. In this paper we provide an alternate approach to prove boundedness of the operator T_{\Omega} for the complete range of exponents \mathcal{H}^{q} for \Omega\in L^{q},\;1<q\leq\infty in dimension one. In particular, we prove the following theorem. Theorem 1.1 . Let 1<q\leq\infty and (p_{1},p_{2},p)\in\mathcal{H}^{q} . Then, for all \Omega\in L^{q}({\mathbb{S}}^{1}) with \int_{{\mathbb{S}}^{1}}\Omega(\theta)d\sigma(\theta)=0 , we have \|T_{\Omega}(f_{1},f_{2})\|_{L^{p}({\mathbb{R}})}\lesssim\|\Omega\|_{L^{q}({\mathbb{S}}^{1})}\|f_{1}\|_{L^{p_{1}}({\mathbb{R}})}\|f_{2}\|_{L^{p_{2}}({\mathbb{R}})}. Our proof of Theorem 1.1 involves a standard spatial localization of the kernel to obtain local kernel of the form K^{0}(y_{1},y_{2})=\Omega\left(\frac{(y_{1},y_{2})}{|(y_{1},y_{2})|}\right){|(y_{1},y_{2})|^{-2}}\beta(|(y_{1},y_{2})|), where \beta\in C_{c}^{\infty}({\mathbb{R}}) with \operatorname{supp}(\beta)\subset(2^{-1},2) . Let T^{0}_{\Omega} be the local operator corresponding to K^{0} defined by T^{0}_{\Omega}(f_{1},f_{2})(x)=\int_{{\mathbb{R}}^{2}}K^{0}(y_{1},y_{2})f_{1}(x-y_{1})f_{2}(x-y_{2})\;dy_{1}dy_{2}. We will prove a trilinear estimate with a decay in the corresponding frequency parameter for the operator T_{\Omega}^{0} acting on functions that are Fourier localized on an annulus. More precisely, we prove the following. Theorem 1.2 . Let 1<q\leq\infty . Then, there exists c>0 such that for all \lambda\geq 1 , we have (1.1) |\langle T^{0}_{\Omega}(f_{1},f_{2}),f_{3}\rangle|\lesssim\lambda^{-c}\|\Omega\|_{L^{q}({\mathbb{S}}^{1})}\|f_{l_{1}}\|_{L^{2}({\mathbb{R}})}\|f_{l_{2}}\|_{L^{2}({\mathbb{R}})}\|f_{l_{3}}\|_{L^{\infty}({\mathbb{R}})}, whenever \operatorname{supp}(\widehat{f_{l_{1}}})\cup\operatorname{supp}(\widehat{f_{l_{2}}})\subset\left\{\lambda\leq|\xi|\leq 2\lambda\right\} and \operatorname{supp}(\widehat{f_{l_{3}}})\subset\left\{|\xi|\leq 2\lambda\right\} for \{l_{1},l_{2},l_{3}\}=\{1,2,3\} . Christ and Zhou [ 3 ] obtained the trilinear smoothing estimates in the context of bilinear maximal functions associated to a certain class of curves in the plane. The proof of Theorem 1.2 is motivated by the ideas developed in [ 3 ] . The key idea in the proof involves a local Fourier series expansion of the test functions. This serves as a substitute to the wavelet decomposition of the multiplier as in [ 14 ] . Further, we perform polar decomposition of the local bilinear operator T_{\Omega}^{0} . A careful oscillatory phase analysis with respect to the spatial and radial variables in the polar decomposition of the operator T_{\Omega}^{0} leads to the trilinear estimate ( 1.1 ). The proof of Theorem 1.2 is carried out in Section 4 . The decay estimate from Theorem 1.2 is the key part in the proof of Theorem 1.1 . Next, we obtain admissible growth estimates for the operator in the respective Banach and non-Banach set of