Ergodic properties of random billiards driven by thermostats (Q2376965): Difference between revisions

The authors analyze a random toroidal billiard model of a many ideal Gibbs distributed particles interacting with a finite amount of disk-shaped thermostats having different temperatures. The resulting discrete-time Markov chain \(\Phi \) possesses on the phase space \(\Omega \) \ an invariant unique ergodic measure absolutely continuous with respect to Lebesgue measure, mixing with exponential rate. The original Markov process also allows such a unique and ergodic measure, whose mixing and convergence of initial distributions do not follow directly because some particles can move very slowly. The main proof of the mixing property is based, in particular, on Harris' Ergodic theorem, as well as on the existence of a special potential \(V\). The following main theorem is proved: {Theorem.} The Markov Chain \(\Phi \) admits a unique absolutely continuous invariant probability measure \(\mu.\) Furthermore, there exists a non-negative function \(V\) on \(\Omega\) and constants \(C>0\) and \( \tilde{\gamma}\in (0,1)\) such that \[ \sup_{A\subset \Omega }|P^{n}((r,v_{\perp }),A)-\mu (A)|\leq C\tilde{\gamma} ^{n}(1+V(r,v_{\perp })). \] Its proof uses a slightly weaker formulation of the well-known Harris' Ergodic theorem, known as Geometric Ergodicity Theorem.