Bilinear multipliers and transference (Q2570962): Difference between revisions

This paper deals with extensions of the classical de Leeuw results on restriction of Fourier multipliers to the setting of bilinear multipliers defined by the condition that the bilinear operator \[ T_m(f,g)(x)=\int_{\mathbb R^2} \hat f(\xi)\hat g(\eta) m(\xi, \eta) e^{2\pi i x(\xi+\eta)} d\xi d\eta, \] extends to a bounded operator from \(L^{p_1}(\mathbb R)\times L^{p_2}(\mathbb R)\) into \(L^{p_3}(\mathbb R)\) where \(1/p_1+1/p_2=1/p_3\). The techniques used here are modifications of those used by de Leeuw in the linear case. One of the main results of this paper shows that if \(m\) is a \((p_1, p_2)\)-bilinear multiplier and \(m\) is a regulated bounded function on \(\mathbb R\times \mathbb R\), then \((m(k,k'))_{k,k'\in\mathbb Z}\) is a \((p_1, p_2)\)-multiplier on \(\mathbb Z\times\mathbb Z\) in the sense that the operator, defined on trigonometric polynomials, \[ T_m(P,Q)(\theta)=\sum_{\mathbb Z^2} \hat f(k)\hat g(k') m(k, k') e^{2\pi i \theta(k+k')}, \] extends to a bounded operator on \(L^{p_1}(\mathbb T)\times L^{p_2}(\mathbb T)\) into \(L^{p_3}(\mathbb T)\).