Multilinear multiplier theorems and applications (Q2420238)

The authors study the boundedness of \(n\)-dimensional \(m\)-linear multiplier operators \[ T_\sigma(f_1,\dots,f_m)(x)=\int_{(\mathbb{R}^n)^m}\widehat{f_1}(\xi_1)\cdots\widehat{f_m}(\xi_m)\sigma(\xi_1,\dots,\xi_m)e^{2\pi ix(\xi_1+\cdots+\xi_m)}d\xi_1\cdots d\xi_m, \] under weaker assumptions in the symbol \(\sigma\) than the previously known results (see, e.g., [\textit{R. R. Coifman} and \textit{Y. Meyer}, Ann. Inst. Fourier 28, No. 3, 177--202 (1978; Zbl 0368.47031)], [\textit{C. E. Kenig} and \textit{E. M. Stein}, Math. Res. Lett. 6, No. 1, 1--15 (1999); erratum ibid. 6, No. 3--4, 467 (1999; Zbl 0952.42005)] and [\textit{L. Grafakos} and \textit{R. H. Torres}, Adv. Math. 165, No. 1, 124--164 (2002; Zbl 1032.42020)]). In particular, their main results is the following: Suppose \(1\le r\le2\) and \(\gamma_{il}>1/r\), for all \(1\le i\le m\) and \(1\le l\le n\). Let \(\sigma\) be a bounded function on \(\mathbb{R}^{mn}\) such that \[ \sup_{j\in\mathbb{Z}}\bigg\|\prod_{i,l}(I-\partial^2_{\xi_{il}})^{\gamma_{il}/2}[\sigma(2^j\cdot)\widehat\Psi]\bigg\|_{L^r(\mathbb{R}^{mn})}=A<\infty, \] where \(\widehat{ \Psi}\) is a smooth function supported in the annulus \(1/2\le|(\xi_1,\dots,\xi_m)|\le 2\) in \(\mathbb{R}^{mn}\) so that \[ \sum_{j\in\mathbb Z}\widehat\Psi(2^{-j}(\xi_1,\dots,\xi_m))=1, \] for all \((\xi_1,\dots,\xi_m)\in\mathbb{R}^{mn}\setminus\{0\}.\) If \(1 < p_i <\infty\), \(i=1,\dots,m\) satisfy that \[ \max_{1\le i\le m}\max_{1\le l\le n}\frac1{\gamma_{il}}<\min_{1\le i\le m} p_i, \] and \(1/p=1/p_1+\cdots+1/p_m\), then we have \[ \|T_\sigma\|_{L^{p_1}(\mathbb{R}^n)\times\cdots\times L^{p_m}(\mathbb{R}^n)}\lesssim A. \] Endpoint estimates on the Hardy space \(H^1(\mathbb{R}^n)\), when all \(p_i=1\), are also considered.