Immunization of geographical networks

Abstract

We numerically investigate the epidemic spread phenomena and efficient immunization strategies on complex networks embedded in geometry. It is assumed that there exists an unavoidable time delay (we call it the detection time) between the actual infection and the beginning of immunization, and we implement two different immunization strategies: one is based on topological connection neighbors (CN) of the infected vertex and the other on geographical spatial neighbors (SN). It is found that the decrease of the detection time is very important for a successful immunization. Our results suggest that within the limitation of the network models considered here, in which the infection probability is assumed to decrease with the geographic distance, the simple SN strategy works almost equally or better than the CN strategy, especially when the detection time is longer. © 2009 ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering.