Time series cross prediction in a single transistor chaotic circuit using neural networks

Abstract

In this paper we will be trying to cross predict a multivariate time series of a single transistor chaotic circuit using neural networks. For this purpose we investigate the influence of a number of first and second order near neighbors in predicting chaotic time series using a back propagation neural network. This influence is examined by changing the number of neurons in the hidden layer of a backprogation neural network with one hidden layer. The number of neurons at the input layer were equal to the embedding dimension of the corresponding strange attractor.

Keywords

• Chaos
• neural networks
• time series prediction
• electronic circuits