A Novel Evolving Network Model with Widely Weighted Dynamics

It is well known that in Barrat-Barthelemy-Vespignani (BBV) model, the rearrangement of weights is local. In this paper, we present a novel weighted evolving network model that allows the flows to be widely updated. This model gives power-law distributions of degree, weight and strength, as confirmed in many real networks. Particularly, the exponents are nonuniversal and depend on a parameter that controls the total weight growth of the network. And it is shown that the droop-head and heavy-tail properties of these distributions, which are observed in many real-world networks, can be reflected by this new network model. It turns out that the strength highly correlates with the degree and displays scale-free property, which is in consistence with empirical evidence. Simulations are provided to demonstrate the theoretical results.