Linearization of generalized interval exchange maps

From MaRDI portal
Publication:1928615



Abstract: A standard interval exchange map is a one-to-one map of the interval which is locally a translation except at finitely many singularities. We define for such maps, in terms of the Rauzy-Veech continuous fraction algorithm, a diophantine arithmetical condition called restricted Roth type which is almost surely satisfied in parameter space. Let T0 be a standard interval exchange map of restricted Roth type, and let r be an integer geq2. We prove that, amongst Cr+3 deformations of T0 which are Cr+3 tangent to T0 at the singularities, those which are conjugated to T0 by a Cr diffeomorphism close to the identity form a C1 submanifold of codimension (g−1)(2r+1)+s. Here, g is the genus and s is the number of marked points of the translation surface obtained by suspension of T0. Both g and s can be computed from the combinatorics of T0.


Let \(I\) be an open bounded interval. A generalized interval exchange map (g.i.e.m.) \(T\) on \(I\) is defined by the following data. Let \({\mathcal A}\) be an alphabet with \(d \geq 2\) symbols. Consider ``top and ``bottom partitions of \(I\) into \(d\) open subintervals indexed by \({\mathcal A}\). More precisely, the top partition is defined by its singularity points \(u_0^t < u_1^t < \cdots < u_d^t\), where \(I=(u_0^t, u_d^t)\), and a bijection \(\pi_t : {\mathcal A} \rightarrow \{ 1,\ldots,d \}\); then \(I_{\alpha}^t = (u_{\pi_t(\alpha) -1}, u_{\pi_t(\alpha)})\) for each \(\alpha \in {\mathcal A}\) and \(I=\sqcup I^t_{\alpha}\) modulo points \(u_j^t\). The definition of the bottom partition \(I=\sqcup I^b_{\alpha}\) is similar, just replace the letter \(t\) by the letter \(b\). It is also required that \(\pi_t^{-1}(\{1,\ldots,k \}) \not= \pi_b^{-1}(\{ 1,\ldots,k \})\) for all \(k=1,\ldots,d-1\). Consider also a \((d \times d)\)-antisymmetric matrix \(\Omega\), whose element \(\Omega_{\alpha\beta}\) equals \((+1)\) if \(\pi_t(\alpha) < \pi_t(\beta)\) and \(\pi_b(\alpha) > \pi_b(\beta)\), equals \((-1)\) if \(\pi_t(\alpha) > \pi_t(\beta)\) and \(\pi_b(\alpha) < \pi_b(\beta)\), and equals \(0\) otherwise. Then the g.i.e.m. \(T\) is defined on \(\sqcup I^t_{\alpha}\) and its restriction on each \(I^t_{\alpha}\) is an orientation-preserving homeomorphism onto the corresponding \(I^b_{\alpha}\). A g.i.e.m. \(T_0\) is ``standard if \(|I^t_{\alpha}|=|I^b_{\alpha}|\) for each \(\alpha \in {\mathcal A}\), and the restriction of \(T_0\) on each \(I^t_{\alpha}\) is a translation. We say that a g.i.e.m. \(T\) is a simple deformation of a standard i.e.m. \(T_0\) if: (i) \(T\) and \(T_0\) have the same discontinuities, (ii) \(T\) and \(T_0\) coincide in the neighborhood of each discontinuity and of the endpoints of \(I\), (iii) \(T\) is a \(C^r\)-diffeomorphism on each \(I^t_{\alpha}\). The main result of the paper is a local conjugacy theorem, which, stated in particular generality (for simple deformations), can be summarized as follows. {Theorem.} For almost all standard i.e.m. \(T_0\) and for any integer \(r \geq 2\) amongst the \(C^{r+3}\)-simple deformations of \(T_0\), those that are \(C^r\)-conjugate to \(T_0\) by a diffeomorphism \(C^r\)-close to the identity form a \(C^1\)-submanifold of codimension \(d^*=d + (r-\tfrac{1}{2})\cdot \text{rank\,} \Omega - 2r\).



Cites work


Cited in
(50)








This page was built for publication: Linearization of generalized interval exchange maps

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1928615)