Research On Radar Echoes Simulation And Feature Detection Methods For Slow Small Target On Time-evolving Sea Surface With Breaking Waves

This thesis focuses on the problem of floating small target detection in a time-evolving rough sea surface with breaking waves,and aims to propose an effective method for simulating echo signals and target detection.To overcome challenges associated with acquiring real sea clutter datasets and high measurement costs,we propose a novel radar echoes simulation method based on the composite electromagnetic(EM)scattering mechanism for small targets on time-evolving sea surfaces with breaking waves.Moreover,by analyzing the differentiated characteristics of sea clutter signals and target echoes in various transformed domains,a small target detector for sea surfaces has been specifically designed,addressing the low performance of traditional coherent statistical detection methods in detecting floating small targets with low Doppler velocities.The research work has important significance in the fields of ocean remote sensing,radar detection,communication,and navigation.The thesis mainly focuses on the following aspects:1.This thesis establishes a time-evolving sea surface model with breaking waves based on the hydrodynamic equations.The method comprehensively considers various time-varying characteristics of parameters,including wind speed,wind direction,wave height,wavelength,positions of breaking waves,and breaking waves periods.By incorporating these parameters,a realistic and complex rough sea surface model is generated,particularly in high sea states where it can capture the phenomenon of wave breaking.It providing an rough sea surface models for electromagnetic scattering calculations and radar echoes simulations.2.This thesis proposes an optimized element-based scattering model that combines the Two-scale Model(TSM)and an improved Shooting and Bouncing Ray(SBR)method for solving the composite scattering problem of target on time-evolving sea surface with breaking waves.The conventional SBR method assumes the sea surface elements as smooth,resulting in inaccurate computation of the coupled scattering between rough sea surfaces and targets.To overcome this limitation,we introduce a rough surface reflection factor to correct the complex reflection coefficients of the sea surface facets.Additionally,message passing interface(MPI)is employed to accelerate the composite scattering algorithm.By utilizing this optimized element-based scattering model,we can establish more accurate and faster composite scattering models for time-evolving sea surface with breaking waves combined with targets.The development of this model lays a solid foundation for radar echoes simulations and Synthetic Aperture Radar(SAR)imaging simulations.3.This thesis presents a groundbreaking approach to simulating the radar echoes of small target on time-evolving sea surface with breaking waves.It takes into account the EM scattering mechanism and incorporates the actual transmitted linear frequency modulation(LFM)signal from the radar.Various characteristics of the sea clutter signal and target echo signal are analyzed,including time-domain characteristics(amplitude distribution,time correlation,spatial correlation,etc.),frequency-domain characteristics(Doppler spectrum,second-order central moments of Doppler,etc.),and time-frequency domain characteristics(information entropy and fractal features based on fractional Fourier transform domain,etc.).The distribution patterns of these characteristics in different transformation domains are thoroughly examined,considering different polarization modes,sea states,and incident pitch angles.This research provides a solid foundation for the design of target detectors based on a comprehensive understanding of the characteristics of sea clutter signals and target echoes signal.4.Based on the differential characteristics of sea clutter signal and target echoes signal in the time-domain,frequency-domain,and time-frequency domain,this thesis designs single-feature detectors in each of these domains.From these detectors,features with exceptional detection performance are carefully selected and integrated to create a multi-transformation domain feature fusion detector.Through a comprehensive evaluation of the detection performance of each detector,the effectiveness of the multi-transformation domain feature fusion detector is convincingly demonstrated.5.This thesis introduces a novel approach for Kelvin wake detection in numerically simulated SAR images using the Cauchy proximal splitting(CPS)algorithm based on the Forward-backward(FB)method.The study explores the impact of oceanic parameters(such as sea conditions),ship motion parameters(such as ship speed,heading,etc.),and SAR system parameters(such as polarization mode,band,etc.)on the detection of ship Kelvin wakes.Moreover,speckle-free SAR images of the ship’s Kelvin wake were combined with gamma noise sequences to simulate the influence of actual jammers on ship Kelvin wake detection.The research conclusions are of value to the development of ship wake stealth technology and the improvement of ship safety.