State Feedback Method to Control Chaotic Neural Network Based on the Dynamic Phase-Space Constraint

Abstract

Chaotic neural networks are versatile systems that attract the attention of researchers while the control of their output is a challenging problem. The objective of this paper is to control chaotic neural networks by a novel combinatorial method adopted from two controlling strategies: the threshold and the damping mechanisms. In one sense, the threshold mechanism restricts the refractoriness internal states with a time varying threshold. The limiting threshold depends on a control signal which is a control signal provided by an inner loop as a function of the network internal state. In another sense, the damping mechanism modifies the network’s dynamics to stabilize the chaotic behaviour basically when the threshold mechanism fails. This mechanism is an outer feedback control loop evaluated when the model is chaotic and exponentially stabilizes it. Two simulation examples are considered in this paper which checks the performance of the proposed method compare to the results of the conventional methods. Comparative results imply the superiority of the proposed controlling method compare to the counterparts on both benchmarks.