Modified SI Epidemic Model for Combating Virus Spread in Spatially Correlated Wireless Sensor Networks

TL;DR

This paper introduces a modified SI epidemic model tailored for wireless sensor networks, accounting for spatial correlation and network parameters, to better understand and control virus spread and network failure.

Contribution

It develops a novel SI model incorporating spatial correlation and sensor network parameters, providing insights into virus dynamics and control strategies in WSNs.

Findings

The model captures spatial and temporal virus spread dynamics.

Control schemes effectively limit infection propagation.

Numerical analysis validates the model's accuracy.

Abstract

In wireless sensor networks (WSNs), main task of each sensor node is to sense the physical activity (i.e., targets or disaster conditions) and then to report it to the control center for further process. For this, sensor nodes are attached with many sensors having ability to measure the environmental information. Spatial correlation between nodes exists in such wireless sensor network based on common sensory coverage and then the redundant data communication is observed. To study virus spreading dynamics in such scenario, a modified SI epidemic model is derived mathematically by incorporating WSN parameters such as spatial correlation, node density, sensing range, transmission range, total sensor nodes etc. The solution for proposed SI model is also determined to study the dynamics with time. Initially, a small number of nodes are attacked by viruses and then virus infection propagates through its neighboring nodes over normal data communication. Since redundant nodes exists in correlated sensor field, virus spread process could be different with different sensory coverage. The proposed SI model captures spatial and temporal dynamics than existing ones which are global. The infection process leads to network failure. By exploiting spatial correlation between nodes, spread control scheme is developed to limit the further infection in the network. Numerical result analysis is provided with comparison for validation.