Method For Complex Network Control Based On Important Nodes

The purpose of complex network research is to understand the topology and function of the network,and finally realize the control of the network.The control of the network is achieved by applying an appropriate control signal to a set of nodes,and how to find the minimum driver node set is a key problem in the field.The minimum control set is composed of nodes,different nodes have different control capabilities,and the control ability of nodes under different network state changes is different.Therefore,how to identify the important control nodes in the network is of great significance for network control.The existing centrality indicator is based on the network topology,which measure the importance of nodes in topology,without considering the specific state changes of networks.At the same time,when searching the minimal driver node set,the controllability of network topology is only considered,and the corresponding control strategy could not be given for the specific state changes.Therefore,it is of great theoretical significance and practical value to find important control nodes in the network as well as to find a specific minimum control set under the specific network state change.In this paper,we study the control problem of complex network in the condition of particular network state changes.Firstly we analysis the important control nodes in complex network,and use the single direct control efficiency to measure the ability of nodes to control complex network under the change of specific network status.Compared to the existing node centrality indicators,single direct control efficiency considering the topological properties of nodes and the specific changes in the network,which can measure the importance of nodes in different network conditions are different.At the same time,we propose the k-walk minimum driver node set finding method based on the important control node,whose minimal control set is searched for the specific state change.In order to verify the effectiveness of our method,experiments were conducted using Gloss TG,Kohonen,Sci Met,and Sma Gri networks.Through experiments,it is found that the nodes with high single direct control efficiency are nodes also with high control centrality,while the nodes with high degree has low single direct control efficient.When using single direct control efficiency to rank nodes,it is found that the disturbance nodes occupy higher ranking in single direct control efficiency.The single direct controlefficiency of nodes has better results under the condition of less disturbance nodes,and the nodes with the best control effect can be found.With the increase of the disturbance nodes,the node control efficiency is reduced.When the network state changes randomly,the node control efficiency is not practical.When the single direct control efficiency is used to find the minimum driver set,the minimum driver set has higher distribution of single direct control efficient than the maximum matching method.The single direct control efficiency and the k-walk minimum driver set finding method based on the important control nodes have good results,and it is of great significance for the actual control of complex network.