Directional complexity and entropy for lift mappings (Q256899): Difference between revisions

Let \(\mathbb{T}=\mathbb{R}/\mathbb{Z}\) be the circle and denote by \(\pi\) the projection \(\mathbb{R}\to\mathbb{T}\). Then each continuous map \(f:\mathbb{T}\to\mathbb{T}\) has a \textit{lift}, which is a continuous map \(F: \mathbb{R}\to\mathbb{R}\) such that \(\pi\circ F=F\circ\pi\) and a degree \(d\), which is an integer \(d\) such that \(F(x+1)=F(x)+d\) for all \(x\in\mathbb{R}\). There are countably many lifts of \(f\), differing by an integer, so we fix one, and note that \(d\) does not depend on the lift. For general continuous circle maps of degree \(1\) we define the \textit{rotation number} of \(x\in\mathbb{T}\) by \[ \rho(F,x)=\lim_{n\to\infty}\frac{F^n(x)-x}{n} \] provided that the above limit exists. The \textit{rotation interval} (perhaps degenerated to a point) is the set of all rotation numbers \(\rho(F,x)\) that do exist. Indeed, this is a closed and connected subset of \([0,1]\). In some sense the rotation interval contains all directions (velocities) in which points of the circle move under \(f\). The authors of the paper under review study the topological entropy of the set of points moving in a prescribed direction (with the fixed rotation number). It turns out that in the setting of degree-\(1\) circle maps this can be seen as a special case of the notion of directional entropy. The authors consider the class of piecewise affine Markov maps of the circle of degree \(1\). For these maps the existence of a Markov partition leads to precise formulas for the directional entropy by reducing the problem to a purely combinatorial one.