This article is cited in
1 paper
Canonical ensemble of particles in a self-avoiding random walk
V. I. Alkhimov Information Technology Faculty, Moscow State University of
Psychology and Education, Moscow, Russia
Abstract:
We consider an ensemble of particles not interacting with each other and randomly walking in the
$d$-dimensional Euclidean space
$\mathbb R^d$. The individual moves of each particle are governed by the same distribution, but after the completion of each such move of a particle, its position in the medium is "marked" as a region in the form of a ball of diameter
$r_0$, which is not available for subsequent visits by this particle. As a result, we obtain the corresponding ensemble in
$\mathbb R^d$ of marked trajectories in each of which the distance between the centers of any pair of these balls is greater than
$r_0$. We describe a method for computing the asymptotic form of the probability density
$W_n(\mathbf r)$ of the distance
$r$ between the centers of the initial and final balls of a trajectory consisting of
$n$ individual moves of a particle of the ensemble. The number
$n$, the trajectory modulus, is a random variable in this model in addition to the distance
$r$. This makes it necessary to determine the distribution of
$n$, for which we use the canonical distribution obtained from the most probable distribution of particles in the ensemble over the moduli of their trajectories. Averaging the density
$W_n(\mathbf r)$ over the canonical distribution of the modulus
$n$ allows finding the asymptotic behavior of the probability density of the distance
$r$ between the ends of the paths of the canonical ensemble of particles in a self-avoiding random walk in
$\mathbb R^d$ for
$2\le d<4$.
Keywords:
canonical ensemble, self-avoiding random walk, constitutive equation,
renormalization group, saddle-point method, asymptotic distribution. Received: 25.01.2016
Revised: 14.02.2016
DOI:
10.4213/tmf9154