An Attack Strategy of Cascading Failures Based on Neighbor-Nodes Average Degree

Recently the attack strategy has become a hot topic to study the robustness against cascading failures in complex networks. Considering a node's initial load and the average degree of the neighboring nodes, we propose a new attack strategy, neighbor-nodes average degree (NAD), and analyze the correlation between the critical threshold TC and tunable parameter α. By comparing the two attack strategies on the nodes with the highest load and the lowest load, we find that the proposed attack strategy is an efficient way to destruct the BA scale-free networks in the range of α. In addition, on the base of the model of BA-BA coupled networks, the preliminary numerical simulation on the proposed attack strategy verifies the mathematic analysis of the targeted-attack to random-attack. By creating a model of the correlated coupled networks for the emergency, we can investigate the condition on cascading failure for the model and propose a protective strategy for preventing avalanche of coupled networks.