Uniform estimates for bilinear Hilbert transforms and bilinear maximal functions associated to polynomials (Q2816468)

Let \(d\) be a positive integer and \(P\) be a polynomial of degree \(d\) given by \(P(t)=a_d t^d+ a_{d-1}t^{d-1}+\dots+a_2 t^2\), where \(a_d,\dots, a_2\in\mathbb R\) and \(a_d\neq0\). Set \(\Gamma_P=(t,P(t))\) and define the bilinear Hilbert transform along the curve \(\Gamma_P\) as \(H_{\Gamma_P}(f,g)=\mathrm{p.v.}\,\int_{\mathbb R}f(x-t)g(x-P(t))\frac{dt}{t}\). Its maximal function is defined as \(M_{\Gamma_P}(f,g)=\sup_{\varepsilon>0}\int_{-\varepsilon}^{\varepsilon}| f(x-t)g(x-P(t))|{dt}\). When \(\Gamma\) is a monomial or a ``non-flat'' curve, \(L^2\times L^2\to L^1\) boundedness of \(H_{\Gamma_P}\) is known.NEWLINENEWLINE In this paper, the authors establish the \(L^p\times L^q\to L^r\) boundedness of \(H_{\Gamma_P}\) and \(M_{\Gamma_P}\), that is: let \(P\) be as above. Then \(H_{\Gamma_P}\) and \(M_{\Gamma_P}\) can be extended to a bounded operator from \(L^p\times L^q\) to \(L^r\) for \(r>\frac{d-1}{d}\), \(p,q>1\) and \(\frac{1}{r}=\frac{1}{p}+\frac{1}{q}\). In addition, the bounds are uniform in a sense that they depend on the degree of \(P\) but are independent of its coefficients.NEWLINENEWLINEThe authors remark that the constraint \(r>\frac{d-1}{d}\) is superfluous if there is some additional convexity condition for the curve. For instance, for \(\Gamma =(t,t^d)\) and \(d\geq2\), they extend the range of \(r\) to \((\frac{1}{2},\infty)\).NEWLINENEWLINEThey discuss also lower bounds on \(r\) in terms of the decay of the sublevel set \(|\{t:|P'(t)-1|<h\}|\).