Sea Clutter Prosessing And Target Detecting Based On Nonlinear Analysis

Sea clutter refers to radar backscatter wave from sea surface, and it seriously interferes in detection performance of targets within sea clutter. So it is great significance to study sea clutter for radar system design, radar signal processing and targets detection within sea clutter. Chaos and Fractal are two branches of nonlinear science, which have comprehensive applications on a great many of research areas. It has been researched how to use the Chaos, Fractal and other nonlinear theories on sea clutter processing and small targets detection in this paper. Theoretical research and experiment research are combined in this dissertation. On the one hand, the lastest nonlinear theories are used to analyze and process sea clutter. On the other hand, IPIX real-life data are used to verify the effectiveness of the theories and methods.The limitation that small and weak targets within sea clutter are difficult to be detected by statistical model and likelihood ratio test is analyzed in this paper. The application of nonextensive distribution model for sea clutter modeling and small targets detecting is improved. Surrogate data method is used to analyze nonlinear character of sea clutter. According to nonlinear characters difference between sea clutter and small targets, small targets detecting method based on surrogate data method is proposed. After nonlinear characters of sea clutter are confirmed, sea clutter is researched by two nonlinear methods of Chaos and Fractal. Firstly, False Nearest Neighbors method is improved by Cao method, and accurate phase reconstruction parameters of sea clutter and targets data are obtained. It is analyzed that sea clutter is a composite of stochastic component and determinate component quantitatively by Cao method. Secondly, through the simulation experiment and real-life data analysis, it is found that the calculation of correlation dimension and largest lyapunov exponent is seriously influenced by noise. And the limitations of judging whether time series has chaotic character or not by the two chaotic invariants are pointed out. At last, fractal theory is used to analyze fractal characters of sea clutter. Targets detection method based on spatial fractal character difference is proposed, which improved the detection performance of small targets within sea clutter.Nonlinear theories are used to analyze the characters of sea clutter, which widens the practical applications of nonlinear theories and deepens the understandings of characters and physical mechanism of sea clutter. The limitations of correlation dimension and largest lyapunov exponent on processing sea clutter time series are researched, which can provide help for other real-life time series nonlinear analysis. According to nonlinear characters difference between sea clutter and small targets, targets detection method of nonextensive model is improved, targets detection methods based on surrogate data and spatial fractal character difference are proposed. All these nonlinear detection methods can improve detection performance of small targets within sea clutter, which no need to add any hardware facility. So this research has important theoretical value and referenced significance in real engineering application.