Effect of Heterogeneous Transmission Rate on Epidemic Spreading Over Scale Free Networks

TL;DR

This study investigates how heterogeneous transmission rates influence epidemic spread over scale-free networks using a Markov chain approach, revealing critical contact thresholds and parameter-dependent epidemic sizes.

Contribution

It introduces a microscopic analysis of epidemic spreading on heterogeneous, scale-free networks considering variable transmission rates and identifies critical contact thresholds for outbreaks.

Findings

Epidemic spread is critical below a minimum contact threshold.

Epidemic size increases with the number of contacts and links per node.

Heterogeneity in transmission rates affects the epidemic dynamics.

Abstract

In the present work the spread of epidemic is studied over complex networks which are characterized by power law degree distribution of links and heterogeneous rate of disease transmission. The random allocation of epidemic transmission rates to the nodes results in the heterogeneity, which in turn causes the segregation of nodes in terms of various sub populations. The aim of the study is to gain microscopic insight into the effect of interactions among various sub populations in the spreading processes of disease over such networks. The discrete time Markov chain method based upon the susceptible infected susceptible (SIS) model of diseases transmission has been used to describe the spreading of epidemic over the networks. The study is parameterized in terms of variable $\lambda$, defined as the number of contacts a node makes with the fraction of its neighboring nodes. From the simulation results, it is found that the spread of epidemic on such networks is critical in terms of number of minimum contacts made by a node below which there is no outbreak of disease. The degree of infection in these networks is assessed from the size of epidemic defined in terms of fraction of infected nodes of the total number and their corresponding level of infection. The results of the parametric study demonstrates the dependence of the epidemic size upon number of concurrent contacts made by a node ($\lambda$ ) and the average number of links per node. In both these cases, the size of the epidemic is found to increase with the corresponding increase in respective parameters.