Research On Radar Multi-component Micro-doppler Signal Decomposition And Parameter Estimation

Recent years,the characteristic analysis and identification of radar targets have attracted more and more researchers' attention.In this domain,the problem of multi-component signals' decomposition and parameter estimation remains a difficult and hot issue to be resolved.This paper has done some research on this issue.In chapter one,we give a description of the background and research values of the target issue.And then the corresponding research status and some classic algorithms are presented.The structure of this paper is presented in the end.In chapter two,we construct the motion model of space precession target at first.After this,the return signal and its modulation effect of both the ideal scattering center model and the anisotropic scattering center model are analyzed,respectively.In chapter three,based on the Fractional Fourier transform and FFT,a multi-component linear frequency modulation signal decomposition method is proposed,which has excellent anti-noise quality.Besides,the proposed method also works for the signal containing weak signal component.In chapter four,the decomposition algorithm based on the inverse Radon transform and the energy concentration measure factor,which is applied to the multi-component SFM signals,is presented and demonstrated.The inverse Radon transform can transform the standard sinusoidal curve to a point in the corresponding domain and reach the goal of energy concentration.When the inverse Radon transform is carried out at the optimal angle,the energy concentration is also optimized.And on the basis of the optimal transform angle,the signal parameter can be estimated.The energy concentration is calculated by the concentration factor.In chapter five,we propose a decomposition algorithm based on the singular value decomposition(SVD)and the K-means clustering,which is suited for signals containing several sinusoidal modulated signal components and at most one non-sinusoidal but periodic modulated signal component.When the original signal is presented in the form of Hankel matrix and is decomposed using the singular value decomposition,the original signal is decomposed into a group of child signal linearly.In chapter six,the main work and the innovation of the paper is summarized.Also,the shortcomings and the forthcoming research direction are pointed out.