Rotation sets of maps of the annulus (Q1091006)

We introduce the rotation set for a degree 1 endomorphism of the annulus. Associated to each orbit we define a real sequence that measures the ''wrapping'' of the orbit around the inner boundary of the annulus. The limit of this sequence, if it exists, is called the rotation number of the orbit. If the limit does not exist, the set of all limit points of the referred sequence is a closed interval and designated rotation interval. The rotation set is the union of all rotation intervals, it is a topologically invariant set that contains its supremum and infimum, however we don't know if it is always closed. The rotation number of a point is defined using the asymptotic average of the radial component of the forward orbit of the point, the \(\omega\)-limit set is defined using the forward asymptotic behavior of a point. One would expect that if a point x has rotation number \(\alpha\) then the rotation set of \(\omega\) (x) is reduced to \(\alpha\). However this is not the case, there are examples of points with rotation number whose rotation set of the \(\omega\)-limit set is nontrivial, but the rotation interval of an orbit is always contained in the convex hull of the rotation set of its \(\omega\)-limit set. There are interesting similarities between properties of the rotation set and topological entropy. Katok proved that any \(C^{1+\epsilon}\) diffeomorphism, f, with positive entropy has an invariant hyperbolic set \(\Lambda\) where f is topologically conjugate to a shift, this implies that either every point in \(\Lambda\) has the same rotation number or there is some point without rotation number. Nevertheless the relation between positive entropy and existence of points without rotation number is not completely understood.