Sea Clutter Modeling And Analysis Based On ARIMA-GARCH Model And Research On SAR-GMTI Technology

Radar plays an important role in communications,meteorology,ranging and navigation.With the in-depth application of radar in both military and civil fields,the design and development of radar have attracted extensive attention from all over the world.Especially in the Marine background,the detection ability of the radar will be affected by the continuous fluctuation characteristics of the ocean surface.In this paper,various statistical models of sea clutter are studied and summarized,and a sea clutter modeling method using summation autoregressive moving mean-generalized autoregressive conditional heteroscedasticity(ARIMA-GARCH)model is proposed,and the mathematical model is applied to the target detection of sea clutter,realizing the first application of ARIMA-GARCH model in the field of sea clutter visual measurement.In addition,in SAR images,the target motion is mainly manifested as range direction diffusion and azimuth direction defocusing,and when the target has velocity component in the range direction,it will lead to azimuth deviation in the range doppler image.Up to now,most of the research on SAR-GMTI subject focuses on clutter suppression.The commonly used method of clutter suppression is multi-channel,because multi-channel can provide more degrees of freedom for target detection.Recently,with the development of computer technology,researchers have gradually turned their attention to single-channel SAR-GMTI technology.This is because a single channel can overcome problem of multi-channel correlation that has been mentioned above.In this paper,three GMTI methods are proposed on the basis of single-channel SAR image,and their feasibility and effectiveness are verified by measured data,and the performance analysis and comparison of three GMTI methods are presented.This paper focuses on the theory and application of sea clutter modeling and ground moving target indication(GMTI)technology in SAR images.The main work of this paper is as follows:1)It briefly introduces six kinds of commonly used clutter statistical characteristics models and statistical analysis methods that have been put forward at present.The commonly used probability density functions are summarized such as Rayleigh distribution,Log-normal distribution,Weibull distribution and K distribution.At the same time,the model of power spectrum is given,the time-dependent characteristics of sea clutter are briefly introduced,and the research and analysis about sea clutter modeling is the main works in this paper.2)The sea clutter based on Log-Normal distribution is described and the simulation results of ARIMA-GARCH model are given.Finally,the sea clutter data are analyzed,and the steps and results of preprocessing,specific analysis and goodness of fit test of the sea clutter data are arranged.By combining the analysis results of the sea clutter data with the above theoretical data,the sea clutter is more well-researched.3)Three moving target detection(MTI)methods based on SAR images are proposed,which are average correlation coefficient method,doppler spectral difference method and doppler center estimation method.The detection results of the three methods are verified by the experimental results of measured radar image data.