Research On Sea Clutter Modeling

Sea clutter is the undesired radar echoes from sea surface.The existence of sea clutter contribute a serious disturbance to the target detection performance of radar. Modeling sea clutter accurately can not only enhance target detection performance ofradar but also provide vivid clutter simulation environment for radar simulator.This thesis researchs sea clutter modeling from three aspects: statistical models,Tsallis distribution model and fractal model.First, coherent Rayleigh, coherented Lognormal, coherent Weibull and Coherent K clutter are simulated with Zero Memory Nonlinearity transform and Spherically Invariant Random Process in the thesis.Through the analysis of IPIX radar real data, it is shown that HH polarization data exhibit a Lognormal behavior, whereas VV polarization data are better fitted by K-distribution.At the same time, correlation characteristic is analyzed by correlation function and power spectrum density. With reference to statistical modeling, autoregressive model is adopted to simulate correlated Gaussian sequences, which is one of our specialities. Order selection is analyzed as an emphasis and a valuable conclusion is gotten.Second, Tsallis distribution model is adopted to implement sea clutter modeling in this thesis.Tsallis distribution can not only fit sea clutter accurately, but also all of the parameters estimated are sensitive to targets.Target detection performance ofβparameter is analysed as an emphasis and target detection base on space Tsallis distribution parameter difference is proposed .At the same time,sea clutter has the power law sensitivity property which is intermediate between stochastic models and chaotic models by analyzing nonextensive parameter q.Finally, fractal characteristic is researched in the thesis. Fractional Brownian motion characteristic of sea clutter is validated and fractal dimention of sea clutter without taget and sea clutter with target are calculated by box-dimention method.Target detection based on fractal dimention is analyzed in brief and sea clutter multi-fractal characteristic is analyzed by MF-DFA. The generated data by statistical models can be used in radar signal simulator. This thesis makes an innovative research to sea clutter modeling and target detecting by Tsallis distribution model and fractal model.