A non-standard discretized SIS model of epidemics (Q2130068)

This paper studies a susceptible-infected-susceptible model with the use of a non-standard discretization method. For a homogeneous continuous-time SIS model, \(S(t)\) is the number of susceptible individuals and \(I(t)\) is the number of infected individuals. The continuous model is given by \(\dot{S}=C-\beta SI+\gamma I-\mu S\) and \(\dot{I}=\beta SI-(\gamma+\alpha+\mu)I\), where \(C\) is an inflow into the population, \(\beta\) is illness transmission rate, \(\gamma\) is recovery rate, \(\mu\) is natural death, and \(\alpha\) is disease induced death rate. Let \(x=\beta S\), \(y=\beta I\), \(k=1+\alpha+\mu\), the system reduces to \(x'=C-xy+y-\mu x\) and \(y'=xy-ky\). The reproduction number is given by \(C/(\mu k)\). With a non-standard discretization method, the model becomes \(x_{n+1}=x_n+h(C-x_ny_n+y_n-\mu x_n)\) and \(y_{n+1}=y_n+h(x_ny_n-ky_n)\), where \(h\) is a step size of discretization. Some basic properties of this discrete-time model have been discussed. The stability analysis of stationary states of the system has been done. Some numerical examples are provided to illustrate the theoretical results.