The Research Of Complex Network Source Location Based On Effective Distance

In real life,many problems can be modeled as source localization of the complex network,such as determining the source of epidemics,looking for the source of rumors,etc.These phenomena tend to produce a cascading effect on the network.Thus,locating information sources quickly and accurately on the network has important significance on infectious diseases control and public opinion control.In this paper,the concept of effective distance is used to measure the close degree between nodes,which can make the topology structure of the complex and difficult to understand.Make the effective distance of the use of network in the infected nodes to disseminate the point source centered a concentric circle.Therefore,the use of"concentric circles" feature as a candidate source to determine whether a point source propagation basis.However,there are two main problems in the existing localization algorithms,one is the need of all nodes of the infected snapshot,the two is not ideal for the positioning accuracy.In view of the above problems,this paper based on the complex network theory,completed the following two aspects of the work.Firstly,the paper proposes a hierarchical localization method based on partial observation.The algorithm adopted in the network deployment observation points to reduce the monitoring cost and computational overhead,and combined with hierarchical localization method to get the localization results of each layer.Finally,According to the position voting system,the final estimate source is obtained.In the model network and real network experiments,the improved algorithm can improve the localization accuracy and improve the efficiency of the localization.Secondly,in order to meet the requirements of fast positioning in some emergency situations,the paper presents a method of real-time location based on effective distance.In the initial stage of infection,the positioning algorithm is executed.And in the followi ng times it would position multiple times,then this algorithm acquires the final estimation of the source point by integrating the multiple positioning results.Compared with the existing algorithms,the improved algorithm can effectively improve the accuracy of location in the fast localization.Through experiment analysis on several model networks and the real network.The improved algorithm fully reflects effectiveness.