K-Coverage: A Monitor Node Selection Algorithm for Diffusion Source Localizations

Abstract

The method of selecting monitor nodes has a direct impact on the accuracy of the infection source
localization, and infection source localization methods that use monitor observation tend to focus
on the source localization itself and ignore the selection of monitor nodes. The biggest problem of
using graph centrality to select monitor nodes is that the distribution of the selected monitor nodes
may be too concentrated, thus affecting the effect of infection source localization. In order to solve
the problem of centralized distribution of monitor nodes, a hierarchical method of selecting monitor
nodes using K-shell is proposed. To further improve the effectiveness of the selection, the overlapping
range of neighbors is introduced into the selection method. Through simulation experiments
on various networks of various sizes, the monitor node selection method can effectively improve the
accuracy of infection source localization.