Quantcast

Scaling Concepts in Graph Thoery: Self-Avoiding Walk on Fractal Complex Networks

Research paper by Yoshihito Hotta

Indexed on: 05 Feb '14Published on: 05 Feb '14Published in: Physics - Statistical Mechanics



Abstract

It was discovered a few years ago that many networks in the real world exhibit self-similarity. A lot of researches on the structures and processes on real and artificial fractal complex networks have been done, drawing an analogy to critical phenomena. However, the non-Markovian dynamics on fractal networks has not been understood well yet. We here study the self-avoiding walk on complex fractal networks through the mapping of the self-avoiding walk to the n-vector model by a generating function formalism. First, we analytically calculate the critical exponent {\nu} and the effective coordination number (the connective constant) by a renormalization-group analysis in various fractal dimensions. We find that the exponent {\nu} is equal to the exponent of displacement, which describes the speed of diffusion in terms of the shortest distance. Second, by obtaining an exact solution, we present an example which supports the well-known conjecture that the universality class of the self-avoiding walk is not determined only by a fractal dimension. Our finding suggests that the scaling theory of polymers can be applied to graphs which lack the Euclidian distance as well. Furthermore, the self-avoiding walk has been exactly solved only on a few lattices embedded in the Euclidian space, but we show that consideration on general graphs can simplify analytic calculations and leads to a better understanding of critical phenomena. The scaling theory of the self-avoiding path will shed light on the relationship between path numeration problems in graph theory and statistical nature of paths.