On the boundedness of the bilinear Hilbert transform along ``non-flat'' smooth curves. The Banach triangle case \((L^r, 1 \leq r < \infty)\) (Q1709563)

Let \(\Gamma=(t, -\gamma(t))\) be a curve such that \(\gamma\) is a suitable smooth, non-flat curve near zero and infinity. The author analyzes the boundedness of the operator \(H_\Gamma\) given by \(H_\Gamma(f,g)(x)=\mathrm{p.v.} \int_{\mathbb R} f(x-t)g(x+\gamma(t))\frac{dt}{t}\) for functions \(f,g\) in the Schwartz class acting from \(L^p(\mathbb R) \times L^p(\mathbb R)\) into \(L^r(\mathbb R)\) for \(1<p<\infty\), \(1<q\leq \infty\) and \(1\leq r<\infty\) with \(1/p+1/q=1/r\). A little history about achievements in this direction: First the case \(\gamma(t)=t\) is the celebrated result, solving the Calderón conjecture, shown by \textit{M. Lacey} and \textit{C. Thiele} [Ann. Math. (2) 149, No. 2, 475--496 (1999; Zbl 0934.42012)] about the boundedness of the bilinear Hilbert transform. The case \(p=q=2\) for \(\gamma(t)=t^d\) with \(d\in \mathbb N\) and \(d\geq 2\) is due to \textit{X. Li} [Anal. PDE 6, No. 1, 197--220 (2013; Zbl 1275.42022)]. The author extended this result for a suitable class of curves, improving the previous case (see \textit{On the boundedness of the bilinear Hilbert transform along ``non-flat'' smooth curves. } Am. J. Math. 137, No. 2, 313-363 (2015)[Zbl 1318.42008].) The case of polynomials of degree \(d\geq 2\) having no constant and no linear terms for values \(1<p,q<\infty\), \(1/p+1/q=1/r\) and \( d/(d-1)<r<\infty\) where later obtained by \textit{X. Li} and \textit{L. Xiao} [Am. J. Math. 138, No. 4, 907--962 (2016; Zbl 1355.42011)]. In this paper, the author extends his previously mentioned result for ``non-flat'' smooth curves (whose definition requires a number of conditions reflecting the smoothness, no critical points and variation, asymptotic behavior and non-flatness near both the origin and infinity) to the range of parameters \((p,q,r)\) mentioned above. The paper follows the ideas from the previous case \(p=q=2\), and his result is optimal up to end-points.