Frequency-hopping Prediction based on the Chaotic Neural Network

Chaos is a nonlinear dynamic behavior exist beyond us. Since the chaotic frequency-hopping code´s characters in chaotic dynamic system, The paper presented a novel chaotic diagonally recurrent neural network approaches to the chaotic fh code prediction, and also presented a momentum gradient backpropagation training algorithm. In order to evaluate the network´s performance of prediction, a simulation approaches to the prediction of fh codes generated by Mackey-Glass chaotic series with this model was carried out. Simulation result indicate that the presented network model can make a rapid and accurate prediction of the chaotic series, and satisfy the multiple step prediction in some degree, which is an effective method approaches to the prediction of Fh-codes.