| Weak target detection in complex sea clutter background is a continuous hot research topic. The traditional target detecting methods are always based on a certain statistic model, whose detection probability depends on the match extent of the sea clutter model and the signal-to-clutter ratio (SCR) of the target. Since it is difficult to find an accurate distribution function to model the complex property of the sea clutter, the detection performance of the traditional methods drops apparently in high state and under low SCR conditions. Since the limitation of the traditional target detecting method, the fractal and multifractal properties of sea clutter and their application on weak target detection are analyzed in this thesis.1.Focused on the performance degradation of the exist fractal based method under low SCR condition, a novel weak target detecting method is proposed based on the fractal properties of AR spectrum in sea clutter background, which extracts a high resolution fractal parameter of AR spectrum and improves the weak target detection performance. Firstly, fractional Brownian motion (FBM) is taken as an example to prove that the fractal property of the power spectrum. To overcome the shortage of Fourier analysis, modern spectral estimation is taken to calculate the power spectrum of sea clutter. Since AR model is a linear predictive model, which estimates the power spectrum of sea clutter form its autocorrelation matrix and has a higher frequency resolution than Fourier analysis, the fractal property of sea clutter is analyzed combined with the AR spectrum. Real measured X-band data is used to verify the fractal property of AR spectrum of sea clutter and AR Hurst exponent and its influencing factors are also analyzed. Finally, AR Hurst exponent is taken as a statstic test to design the CFAR target detection method. The results show that the Hurst exponent of AR spectrum is effective for weak target detection in sea clutter background. Compared to the frequency Hurst exponent based method and the traditional Constant False Alarm Rate (CFAR) method, the proposed method has a better detection performance.2.The multifractal spectrum is an important parameter to describe the multifractal properties of sea clutter. Focused on the nonstability of the target detecting method based on global multifractal spectrum in time domain, a novel weak target detecting method is proposed based on local-multifractal spectrum, which improves the weak target detection performance. Firstly, the three essential factors, non-Gaussian property, long-time correlated property and scale-invariant property, are analyzed to judge the multifractal properties of sea clutter. Then, the multifractal spectrum is calculated by Legendre transformation method, and the relationship between the multifractal spectrum function and the singularity function is analyzed in power distribution aspect. According to the difference of local multifractal spectrum between sea clutter with and without target, we add a rectangle window to the multifractal spectrum and extract the local mean square summation (LMSS) of the multifractal spectrum. In addition, the window length is adaptively computed by the slope of the singularity intensity function. Finally, the experimental results of real S-band and X-band sea clutter data show that the proposed method is effective. In comparison with the existing multifractal spectrum based method and the traditional CFAR method, the proposed method could achieve a better detection performance under low SCR conditions.3.Focused on the lower spectrum resolution of the target detecting method based on multifractal property in frequency domain, a novel weak target detecting method is proposed based on the multifractal properties of AR spectrum in sea clutter background, which increases the power spectrum resolution and improves the weak target detection performance. Firstly, the multifractal theory is introduced into the modern spectral estimate field, and the AR spectral estimation is taken to calculate the power spectrum of sea clutter. With experiment of the real measured X-band radar sea cluter data,the multifractal property of AR spectrum is proved by MF-DFA method, while the AR generalized Hurst exponent in the non-scale interval is aslo calculated. Finally, the AR generalized Hurst exponent and its influencing factors are analyzed. In addition, we extract the local integration of AR generalized Hurst exponent from its spectial interval, which is taken as a statistic test to design CFAR detection method. The experimental results show that the proposed method has the ability to detect weak targets in sea clutter background. Compared to the existing multifractal based method and the traditional CFAR method, the proposed method has a better detection performance.4.Focused on the performance degradation of the exist extended fractal based method under low SCR condition, a novel weak target detecting method is proposed based on the extended fractal properties of AR spectrum in sea clutter background, which increases the accucacy of the extended fractal parameter and improves the weak target detection performance. To overcome the shortage of the Fourier analysis, AR spectral estimation is introduced into the extended fractal analysis of sea clutter. As an extension of the fractal and multifractal anaylsis of AR spectrum, the extended fractal method can analyze the fractal features in different scale. With experiment of the real measured X-band radar sea cluter data, the multi-scale Hurst exponent of AR spectrum and its optimal scale inveral are analyzed. Finally, factors influencing the AR multi-scale Hurst exponent are analyzed, and the AR multi-scale Hurst exponent from its optimal interval is extracted for target detection. The experimental results show that the proposed method has the ability to detect weak targets in se clutter background. Compared to the existing extended fractal based method and the traditional CFAR method, the proposed method has a better detection performance.
|
PDF Full Text Request