Transference of bilinear multiplier operators on Lorentz spaces (Q1419656)

Let \(m(\xi,\eta)\) be a bounded continuous function in \(\mathbb{R}\times \mathbb{R}\). The bilinear multiplier operator \(C_m\) is defined on the product test function space \({\mathcal S}(\mathbb{R})\times{\mathcal S}(\mathbb{R})\) by \[ C_m(f,g)(x)= \int_{\mathbb{R}} \int_{\mathbb{R}} \widehat f(\xi)\widehat g(\eta) m(\xi,\eta) e^{2\pi ix(\xi+\eta)} d\xi\,d\eta. \] Similarly, for \(0< \varepsilon< \infty\), the family of bilinear multiplier operators \(T_{m,\varepsilon}\) is defined on the product test function space \({\mathcal S}(\mathbb{T})\times{\mathcal S}(\mathbb{T})\) by \[ T_{m,\varepsilon}(f,g)(\theta)= \sum_k \sum_j \widehat f(k)\widehat g(j) m(\varepsilon k,\varepsilon j) e^{2\pi i\theta(k+ j)}. \] In a previous paper [``Transference on certain multilinear multiplier operators'', J. Aust. Math. Soc. 70, No. 1, 37--55 (2001; Zbl 0984.42006)] \textit{D. Fan} and \textit{S. Sato} established a de Leeuw type theorem which says that \(C_m\) is bounded from \(L^p(\mathbb{R})\times L^q(\mathbb{R})\) to \(L^r(\mathbb{R})\) if and only if \(T_{m,\varepsilon}\) are bounded from \(L^p(\mathbb{T})\times L^q(\mathbb{T})\) to \(L^r(\mathbb{T})\) uniformly for \(\varepsilon> 0\), where \(1/p= 1/q= 1/r\). In the paper under review, the authors extend the above de Leeuw type theorem to Lorentz spaces.