Novel results on global stability analysis for multiple time-delayed BAM neural networks under parameter uncertainties

Abstract

This paper describes a new global robust stability analysis of bidirectional associative memory (BAM) neural networks. Under parameter uncertainty, we find a new upper bound on the norm of the weight matrix of the synaptic connection of time-delayed BAM neural networks. Our new upper bound provides a different kind of sufficient condition for the equilibrium point pertaining to the robust stability of BAM neural networks. Several classes of bounded activation functions are formulated. In addition, appropriate Lyapunov–Krasovskii functional (LKF) candidates are used in the process of deriving the new sufficient conditions for the BAM neural networks that are independent of the time delay parameters. We conduct some comparative studies with numerical examples to demonstrate the advantages of our findings over the stability results in terms of BAM neural network parameters.