Rewiring hierarchical scale-free networks: Influence on synchronizability and topology

¹ Research Domain IV - Transdisciplinary Concepts & Methods, Potsdam Institute for Climate Impact Research 14473 Potsdam, Germany
² Department of Physics, Humboldt University of Berlin - 12489 Berlin, Germany

Received: 13 July 2017
Accepted: 18 September 2017

Abstract

Many real-world complex networks simultaneously exhibit topological features of scale-free behaviour and hierarchical organization. In this regard, deterministic scale-free (DSF) (Barabási A.-L. et al., Physica A, 299 (2001) 559) and pseudofractal scale-free (PSF) (Dorogovtsev S. N. et al., Phys. Rev. E, 65 (2002) 066122) networks constitute notable models which simultaneously incorporate the aforementioned properties. The rules governing the formation of such networks are completely deterministic. However, real-world networks are presumably neither completely deterministic nor perfectly hierarchical. Therefore, we suggest here initially perfectly hierarchical scale-free networks with subsequently randomly rewired edges as better representatives of practical networked systems. In particular, we preserve the scale-free degree distribution of the deterministic networks but successively relax the hierarchical structure while rewiring them. We utilize the framework of master stability function in investigating the synchronizability of dynamical systems coupled on such rewired networks. Interestingly, this reveals that the process of rewiring is capable of significantly enhancing, as well as, deteriorating the synchronizability of the resulting networks. We investigate the influence of rewiring edges on the topological properties of the rewired networks and, in turn, their relation to the synchronizability of the respective topologies. Finally, we compare the synchronizability of DSF and PSF networks with that of random scale-free networks (generated using the classical Barabási-Albert (BA) model). We find that the BA random scale-free networks promote synchronizability better than the rewired versions of their deterministic counterparts of DSF and PSF networks.