SEIRS model with spatial correlation for analyzing dynamic of virus spreading in event-driven wireless sensor networks

Abstract

In Event-driven wireless systems, mostly data transmission depends on events occurring in the sensor field. Most of the time, sensor nodes are silent or in sleep mode. When events occur in the sensor field, a single event can trigger many nodes for data transmission. In such a scenario, the nodes collect the correlated information due to the overlapped coverage area. Existing epidemiological designs do not consider the nodes' behavior to investigate infection dynamics for this scenario. In this paper, a susceptible-exposed-infectious-recovered-susceptible (SEIRS) is designed by considering the spatial correlation for analyzing the dynamics of the virus spreading in eventdriven wireless systems. Firstly, we show how strongly correlated nodes and less correlated nodes are formed in a WSN based on sensor coverage. The differential equations of SEIRS are then derived. An analysis on system stability is performed for finding the basic reproduction number R 0 . The value of R 0 gives important significance in terms of spatial correlation for analyzing virus spreading.Experiments are performed to validate the model using various parameters such as correlation, node density, the basic reproduction number. Comparisons with existing models show the effectiveness of the SEIRS model. Based on the analysis, it is observed that the virus spread control can be possible by reducing R 0 . It is also found that the threshold of virus propagation is strongly dependent on the spatial correlation between nodes in the network. The virus is the network persists at virus-free equilibrium when R 0 [ 1 with higher spatial correlation, whereas it becomes globally stable for R 0 \1.