Fourier Analysis on $\mathbb{T}^m\times\mathbb{R}^n$ and Applications to Global Hypoellipticity

Abstract

This article presents a convenient approach to Fourier analysis for the investigation of functions and distributions defined in $\mathbb{T}^m \times \mathbb{R}^n$. Our approach involves the utilization of a mixed Fourier transform, incorporating both partial Fourier series on the torus for the initial variables and partial Fourier transform in Euclidean space for the remaining variables. By examining the behaviour of the mixed Fourier coefficients, we achieve a comprehensive characterization of the spaces of fast decaying smooth functions and distributions in this context. Additionally, we apply our results to derive necessary and sufficient conditions for the Schwartz global hypoellipticity of a class of differential operators defined on $\mathbb{T} \times \mathbb{R}$, including all constant coefficient first order differential operators.