An Ising Model Approach to Malware Epidemiology

TL;DR

This paper applies an Ising model framework to analyze malware spread in networks, linking infection dynamics to network traffic and congestion levels, providing insights into how network conditions influence malware propagation.

Contribution

It introduces an Ising model-based approach to model malware spread, connecting network traffic and congestion to infection rates, which is a novel application in epidemiological modeling of malware.

Findings

Faster malware spread in congested networks

Slower infection rates in more efficient networks

Propagation rate increases with data traffic

Abstract

We introduce an Ising approach to study the spread of malware. The Ising spins up and down are used to represent two states--online and offline--of the nodes in the network. Malware is allowed to propagate amongst online nodes and the rate of propagation was found to increase with data traffic. For a more efficient network, the spread of infection is much slower; while for a congested network, infection spreads quickly.