Capacity of stable periodic solutions in discrete-time bidirectional associative memory neural networks

The existence, stability, and even the number (capacity) of stable periodic solutions in discrete-time bidirectional associative memory neural networks are investigated in this paper. Some fundamental mathematical techniques instead of the bifurcation method, are employed to prove the existence of periodic solutions and establish the relation of the number of periodic solutions and the magnitude of delays. Our results show that we can simply design a delayed neural network model to store memories or patterns as stable periodic solutions.