Random Iteration of Maps on a Cylinder and diffusive behavior
In this paper we propose a model of random compositions of cylinder maps, which in the simplified form is as follows: (θ,r)∈T×R=A and \begin{eqnarray} \nonumber f_{\pm 1}: \left(\begin{array}{c}θ\\r\end{array}\right) & \longmapsto & \left(\begin{array}{c}θ+r+\varepsilon u_{\pm 1}(θ,r). \\ r+\varepsilon v_{\pm 1}(θ,r). \end{array}\right), \end{eqnarray} where u± and v± are smooth and v± are trigonometric polynomials in θ such that ∫v±(θ,r)dθ=0 for each r. We study the random compositions (θn,rn)=fωn−1∘⋯∘fω0(θ0,r0) with ωk∈{−1,1} with equal probabilities. We show that under non-degeneracy hypothesis for n∼ε−2 the distributions of rn−r0 weakly converge to a diffusion process with explicitly computable drift and variance. In the case of random iteration of the standard maps \begin{eqnarray} \nonumber f_{\pm 1}: \left(\begin{array}{c}θ\\r\end{array}\right) & \longmapsto & \left(\begin{array}{c}θ+r+\varepsilon v_{\pm 1}(θ). \\ r+\varepsilon v_{\pm 1}(θ) \end{array}\right), \end{eqnarray} where v± are trigonometric polynomials such that ∫v±(θ)dθ=0 we prove a vertical central limit theorem. Namely, for n∼ε−2 the distributions of rn−r0 weakly converge to a normal distribution N(0,σ2) for σ2=41∫(v+(θ)−v−(θ))2dθ. Such random models arise as a restrictions to a Normally Hyperbolic Invariant Lamination for a Hamiltonian flow of the generalized example of Arnold. We expect that this mechanism of stochasticity sheds some light on formation of diffusive behaviour at resonances of nearly integrable Hamiltonian systems.
