Random Iteration of Maps on a Cylinder and diffusive behavior

Publication date 14 January 2015

Generation, random and stochastic difference and differential equations (37H10) Stability problems for finite-dimensional Hamiltonian and Lagrangian systems (37J25) Perturbations of finite-dimensional Hamiltonian systems, normal forms, small divisors, KAM theory, Arnol'd diffusion (37J40) Stationary stochastic processes (60G10)

Abstract: In this paper we propose a model of random compositions of cylinder maps, which in the simplified form is as follows:

(h e t a, r) i n m a t h b b T i m e s m a t h b b R = m a t h b b A

and �egin{eqnarray}

onumber f_{pm 1}: left(�egin{array}{c} heta\rend{array} ight) & longmapsto & left(�egin{array}{c} heta+r+varepsilon u_{pm 1}( heta,r). \ r+varepsilon v_{pm 1}( heta,r). end{array} ight), end{eqnarray} where $u_{p} m$ and $v_{p} m$ are smooth and $v_{p} m$ are trigonometric polynomials in $h e t a$ such that $i n t v_{p} m (h e t a, r), d h e t a = 0$ for each $r$ . We study the random compositions ( heta_n,r_n)=f_{omega_{n-1}}circ dots circ f_{omega_0}( heta_0,r_0) with $o m e g a_{k} i n - 1, 1$ with equal probabilities. We show that under non-degeneracy hypothesis for $n s i m v a r e p s i l o n^{- 2}$ the distributions of $r_{n} - r_{0}$ weakly converge to a diffusion process with explicitly computable drift and variance. In the case of random iteration of the standard maps �egin{eqnarray}

onumber f_{pm 1}: left(�egin{array}{c} heta\rend{array} ight) & longmapsto & left(�egin{array}{c} heta+r+varepsilon v_{pm 1}( heta). \ r+varepsilon v_{pm 1}( heta) end{array} ight), end{eqnarray} where

v_{p} m

are trigonometric polynomials such that

i n t v_{p} m (h e t a), d h e t a = 0

we prove a vertical central limit theorem. Namely, for

n s i m v a r e p s i l o n^{- 2}

the distributions of

r_{n} - r_{0}

weakly converge to a normal distribution

m a t h c a l N (0, s i g m a^{2})

for

s i g m a^{2} = f r a c 14 i n t (v_{+} (h e t a) - v_{-} (h e t a))^{2}, d h e t a

. 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.

This page was built for publication: Random Iteration of Maps on a Cylinder and diffusive behavior

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6258111)