Analysis of asymmetric neural networks with time delay and its applications in associative memories

Though there were some papers studying stability of neural networks (NNs) with time delay for associative memories, not any paper has studied it by m energy functions method, which can simplify the stable conditions of NNs in practical applications. In this paper, asymmetric NNs with time delay are studied, and one sufficient condition for the global and local stability asymptotic stability of the NNs is given by m energy functions method. A design procedure of NN for associative memories (AMs) is given by using of single value decomposition. Finally, numerical simulation result shows that AMS can be realized by our designing asymmetric NN.