Distributional chaos via isolating segments (Q2467052)

For a compact metric space \((X,\rho)\), a continuous map \(f\colon X\to Y\), a pair of points \((x,y)\in X\), a real number \(t>0\), and a positive integer \(n\) put \[ \xi(x,y,t,n):=\#\{i\colon 0\leq i<n, \rho(f^i(x),f^i(y))<t\}. \] Define \[ F_{xy}(t):=\liminf_{n\to \infty} \xi(x,y,t,n)/n\text{ and }F^\ast_{xy}(t):=\limsup_{n\to \infty} \xi(x,y,t,n)/n. \] The map \(f\) is called distributionally chaotic if there exists an uncountable subset of \(X\) such that for each pair \((x,y)\) of its distinct points, \(F_{xy}(s)=0\) for some \(s>0\) and \(F^\ast_{xy}(t)=1\) for all \(t>0\). If, moreover, \(s\) can be chosen the same for all such pairs, \(f\) is called uniformly distributionally chaotic. Let \(T>0\). A nonautonomous differential equation \[ \dot x=v(t,x), \] where \(v\) is \(T\)-periodic in \(t\), is called (uniformly) distributionally chaotic if its time-\(T\) Poincaré map restricted to some compact invariant set is (uniformly) distributionally chaotic. The main result of the paper asserts that a planar equation \[ \dot z=(1+\exp(i\kappa t)| z| ^2)\bar z \] is uniformly distributionally chaotic provided \(0<\kappa\leq 0.5044\). Its proof is based on the fixed point index theory and some geometric considerations.