Memristor-based Chaotic Neural Networks And Its Applications

Artificial neural network short for neural network is a kind of mathematical model which imitates biological neural network’s structure and function. It consists of multiple interconnected artificial neurons. Neural network can show complex dynamic characteristics via the interaction between neurons, and it is a highly nonlinear dynamic system with the parallel processing ability, information storage capacity, fault tolerance, self-organization and adaptive ability. Chaos is a widespread nonlinear phenomenon with complex and seemingly random behavior. Biological experiments indicate that chaotic behaviors exist in biological neural system, so neural network and chaos are closely related. Chaotic neural network can be implemented with the combination of chaos and neural network. Chaotic neural network is considered to be one of the intelligent information processing systems that can realize the calculation of the real world, it is widely used in optimization, pattern recognition, image processing, associative memory and so on.In this paper, the theoretical knowledge of chaotic neural network has been studied in depth. At the same time, chaos theory has been introduced in detail including the proposing of chaos, the characteristics of chaos and the combination of chaos and neural network. Then some applications of chaotic neural network in associative memory have been described. Moreover, on the basis of the study of memristor which is believed to be the fourth fundamental circuit element, a memristor-based chaotic neural network model has been proposed by using memristor realize the accumulation of the network. A series of computer simulations verify the effectiveness of the proposed scheme. The proposed scheme is expected to greatly simplify the structure of chaotic neural network and is in favour of the circuit realization of chaotic neural network. Besides. Memristor is very suitable for electronic synapse owning to its unique features. Several kinds of synapse circuits have been presented, a memristor-based synapse circuit has been proposed on the basis of the study of amplifier circuit. Further, the spintronic memristor model has been used in the bridge synapses. It is believed that the bridge synapses would have a better performance with the spintronic memristor. The associative memory of chaotic neural network with the bridge synapses has been studied in the end. The network can be controlled with the delay feedback signal, and the output of the network will be cyclical. The research results of this paper will promote the development of chaotic neural network and experiments basis for its physical implementation, which is expected to simplify the structure of chaotic neural network.