The Study Of HF Radar Waves Echo Characteristics Based On Chaos Theory

A long time, sea clutter has always been a concern of the majority of scholars.However, the results of study aren’t well when we have seemed sea clutter as arandom noise. In order to learn further sea clutter inherent physical properties andlaws, the research of sea clutter using chaos theory is a new direction. Under thebackground, we study the chaotic characteristics of HF radar waves echo in thispaper, expecting the application of chaos theory to the research of HF radar wavesecho.In order to apply chaos theory to study the HF radar waves echo, firstly, it is toprove that it has chaotic characteristics. On the basis of the data of HF radar wavesecho, we reconstruct its phase space, including two aspects, which are the delaytime and embedding dimension calculation. The former uses an improvedautocorrelation function method, and the latter uses Cao method. Then thereconstruction works well. Based on the phase space reconstruction, three ways thatare correlation dimension, the Lyapunov exponent and Kolmogorov entropy, provethat HF radar waves echo is chaotic indeed. In the correlation dimension calculation,the author proposes a new and more effective method to identify the scale-free,which can be accurate to calculate the correlation dimension.Secondly, we use the butterfly effect in chaos theory, which is short-termpredictability and long-term unpredictability to predict the HF radar waves echo.Due to the fact that HF radar waves echo is nonlinear and chaotic, so we usenonlinear neural network prediction method, including BP network, RBF network,normalized RBF network and the RBF-BP neural network. The characteristics offour kinds of network are different, but the final prediction performance is similar.In view of these features that the BP network is easy to fall into local minima, theprediction time of normalized RBF network is a little longer, and RBF-BP networkis slightly complex, the simulation results show that the RBF network seems to bebetter.Finally, on the based of sea clutter prediction, the clutter cancellation and smalltarget detection are discussed using chaotic time series and the HF radar waves echo data. The BP network and RBF network are applied to clutter cancellation. Afterthat, the original targets submerged in the sea clutter can be appearance. Then it istarget detection according to the residual error after the clutter cancellation.Simulation experiments show that the target that SCR is lower can be detected well.In the last chapter, several common methods used denoising are briefly researched,and the analysis results show that wavelet denoising method has obviousadvantages.