Forecasting method of chaotic time series based on RBF

The method to analyze sea clutter based on chaos theory is presented. Based on Takens embedding theorem,Chaotic characteristic invariants of the sea clutter data from the IPIX radar of McMaster University,Canada,are computed and analyzed.A chaos based-detector is designed.The detector utilizes radial basis function neural network to reconstruct the chaotic dynamic system,namely, to build a precise prediction model.The simulation has proven its effect.