Random walks in a one-dimensional Lévy random environment (Q288192)

The authors consider a generalization of the so-called Lévy-Lorentz gas, introduced by \textit{E. Barkai, V. Fleurov} and \textit{J. Klafter}, [``One-dimensional stochastic Lévy-Lorentz gas'', Phys. Rev. E, 61, 1164 (2000)], which describes a continuous motion of a particle on the real line between randomly allocated marked points. The distances between consecutive marked points are i.i.d. with finite mean (but possibly infinite variance), and there is a marked point at the origin. The particle starts at the marked point at the origin, chooses an integer \(n\) independently at random according to some symmetric distribution with finite variance, and moves continuously at speed one to the \(|n|\)-th marked point to the right (if \(n>0\)) or to the left (if \(n<0\)) from the current location. Once the particle reaches the target, it selects the next marked point independently by the same rule, and moves towards it continuously with speed one. The procedure repeats indefinitely. The authors prove that the described process satisfies the quenched and annealed central limit theorem, with an explicit variance of the limiting normal distribution. It is conjectured, based on simulations, that for some distributions of the marked points, the annealed second moment of the particle location grows faster than linear. A related discrete time random walk on marked points is also studied. Here, the particle, after identifying the next target marked point by the rules described above, jumps to it instantaneously. The quenched and annealed central limit theorem for this random walk is proven as well as the convergence of rescaled moments of the displacement.