Discrete-time analogues of integrodifferential equations modelling bidirectional neural networks (Q5957939)

The authors formulate discrete-time analogues of integrodifferential equations modelling bidirectional neural networks whose continuous-time version has been studied in the literature. They show that the discrete-time analogues preserve the equilibria of the continuous-time networks. By constructing a Lyapunov-type sequence, sufficient conditions are obtained so that every solution of the discrete-time analogue converges exponentially to the unique equilibrium. These sufficient conditions are identical to those obtained for the continuous-time networks. By constructing discrete analogue of Halanay-type inequalities, the authors obtain another set of sufficient conditions for the global exponential stability of the unique equilibrium of the discrete-time analogue. These conditions have not been obtained in the case of continuous-time networks. Finally, several computer simulations are included to illustrate the advantages of discrete-time analogue in numerically simulating the continuous-time network with distributed delays over finite intervals.