Chaos control and associative memory of time-delay symmetric globally coupled neural network

A time-delay symmetric globally coupled neural network is considered. Firstly, its rich dynamic behaviors and the characteristics variation law dependent on time delay are exhibited. Secondly, by using a parameter control method, the output can be stabilized to corresponding periodic orbits when only partial neurons enter their own periodic orbits. At last, the capacity and the restoration of the dynamic associative memory are shown. The experimental results suggest that it takes shorter time and less energy to implement associative memory by using the stable output control than by using other control methods. Moreover, the large capacity of associative memory and strong anti-interference ability are also demonstrated.