Global asymptotical stability analysis of a class of discrete-time BAM neural networks: An LMI approach

A novel neural network model named standard neural network model ( SNNM ) was advanced. The conditions for the global asymptotical stability of the equilibrium points of the SNNMs were derived, which were formulated as some linear matrix inequalities (LMIs). By using state affine transformation, the extended discrete-time bidirectional associative memory (BAM) neural networks were converted into the SNNMs. Some criteria of global asymptotic stability for the discrete-time BAM neural networks were derived from studies on the SNNMsJ stability. The proposed approach extends the known stability results. Furthermore, this approach, which has less conservative, is easy to implement and applicable to other recurrent neural networks.