Study On Fractal Method For Radar Signal Detection

As a mathematic tool for irregular geometry in the natural world, fractal theory has been founding wide application in such fields as image processing, natural surface modeling, time series analysis etc. successfully. Since the presence of fractal processes and phenomenon in radar system, the new idea and method can be obtained by applying the fractal theory to study and analyze radar signal processing system. The application of the random fractal theory in the field of radar signal detection both theoretically and experimentally, and some methods for modeling & simulation and target detection based on the random fractal model for the clutter from natural rough surface are proposed. The main research work and result is presented as follow: 1 .The random fractal signal is described and analyzed in the respect of fractal basic quantities, fractional integral model, and wavelet model, and the corresponding methods for parameter estimation are proposed. The experimental and theoretical analysis demonstrates that all the methods are available. The fractional differenced method based on the fractional integral model runs fast and the Maximum Likelihood Estimation based on the wavelet model has strong rubbish, where the former is fit for real-time estimation and the latter for estimation under low signal- noise-ratio. 2.The fractal property of the clutter from natural rough surface is studied on both the real-life data and the generation mechanism. On the one hand the analysis on fractal properties of the real-life clutter data is proceeded, and the conclusion that the clutter from natural rough surface has random fractal characteristic is obtained. On the other hand the mechanism of the fractal property is discussed by means of study on Kirchoff solution to the electromagnetic scattering field from both time- variant fractal surface and natural rough surface, and the random fractal model for the clutter from natural rough surface is proposed further. The conclusion that both experimental analysis and theoretically deduction demonstrate that the clutter from natural rough surface do have random fractal feature provides a new way to study the property of the clutter. 3 .The statistic and fractal characteristic of the parameters in the random fractal model for the clutter from natural rough surface is advanced. By applying theeither radix orthogonal wavelet transform, not only the validity of the random fractal model for the clutter is proved, but also the mathematic expressions for the parameters in the model is provided. Combining the traditional idea and method for clutter simulation, the simulation method based on random fractal model is proposed by using X radix wavelet and random Weierstrass function respectively, which is the entia of the statistic property and the fractal property. 4.Combining the theory on model matching and correlation detection, the methods for radar target detection based on the fractal model are proposed in three aspects of multi-fractal spectrum, iterated function system and wavelet whitening filter according to the fact that the clutter from natural rough surface has random fractal feature. Experiment results demonstrate the detection ability of all proposed methods is better. All that has been done provide new idea and method for target detection with strong clutter.