On the microlocal regularity of the analytic vectors for ``sums of squares'' of vector fields (Q2098205): Difference between revisions

Let the linear partial differential operator \[ P(x,D)=\sum_{j=1}^{m}X_{j}^{2}(x,D) \] be a sum of squares of the vector fields \(X_{j}(x,D)\) with real-valued real analytic coefficients on the open neighborhood \(U\) of the origin of \(\mathbb{R }^{n}\). It is assumed that \(P(x,D)\)\ satisfies the classical well-known Hörmander's condition. Denote by \(\mathcal{A}(U,P)\)\ the space of analytic vectors of the operator \( P(x,D)\), and by \(WF_{s}(u),s\geq 1\), the \(s\)-Gevrey wave front of the distribution \(u\). The main aim of this paper is the proof of the following theorem on Gevrey microlocal regularity of the analytic vectors of the operator \(P(x,D)\). The proof is done via FBI transform and the corresponding microlocal analysis. Theorem. Let \(u\in \mathcal{A}(U,P)\), if \((x_{0},\xi _{0})\)\ is a point of the characteristic set of \(P(x,D)\), then \((x_{0},\xi _{0})\notin WF_{\nu (x_{0},\xi _{0})}(u)\). The number \(\nu (x_{0},\xi _{0})\)\ is defined as the length of the shortest iterated Poisson bracket of the symbols of the vector fields which is non zero at \((x_{0},\xi _{0})\).