Research On Intra-pulse Analysis Method Of Radar Signal Based On Time-frequency Transform

With the rapid development of electronics and radar technology,more and more new types of complex systems have become radars,and they have occupied a dominant position.The traditional five-parameter analysis has failed to meet the needs of modern electronic detection,and it needs to extract more stable and fine features in the pulse.To meet the needs of investigation and analysis.The purpose of intra-pulse analysis is to estimate the intra-pulse modulation parameters and identify the intra-pulse modulation method.The selection of modulation and modulation parameters is closely related to the function and purpose of the radar.Therefore,the modulation parameters of the intercepted radar signal are performed.It is of great significance that effective estimation can accurately and effectively evaluate the radar.The electromagnetic environment of modern electronic battlefields has become increasingly complex.Electronic reconnaissance equipment is very dense.The complex and dense electromagnetic signal environment makes the signals intercepted by receivers a mixed signal where multiple radar emitter signals overlap or overlap.Multi-component radar signals,which greatly increase the difficulty of separating and modulating radar signal estimates.Multi-component radar signals are generally difficult to separate directly from the time domain or frequency domain.The timefrequency analysis method is mainly used to extract time-frequency features to achieve separation and parameter estimation.How to effectively separate and extract parameters of multi-component radar signals is a problem worthy of study.In this dissertation,Hilbert-Huang transform is used to select the Hilbert-Huang transform in time-frequency analysis based on time-frequency transform to analyze the multi-component radar signals in the pulse,obtain the Hilbert spectrum real-time frequency distribution of multi-component radar signals,and perform the timefrequency characteristics of the multi-component radar signals.extract.An improved Hilbert-Huang transform combining wavelet packet decomposition and screening process is proposed.The improved algorithm improves the anti-noise performance of the algorithm,effectively suppresses modal aliasing during EMD(Empirical Mode Decomposition)and rejects false frequency components.The SNR is 0 d B.Under the conditions,the noise can be effectively suppressed,and the time-frequency distribution of a correct and clear multi-component radar signal,ie Hilbert spectrum,can be obtained.In order to make the time-frequency lines of radar signals of different components in the time-frequency spectrum more clustered,the method of image processing is then introduced.The Gaussian smoothing filter is used to smooth the Hilbert spectrum to obtain higher time-frequency aggregation,which is a multicomponent radar.The component signals of the signal are prepared for separation and extraction and estimation of the relevant modulation parameters.Compared with the classical WVD(Wigner-Ville Distrubution),SPWVD(Smooth Pseudo Wigner-Ville Distribution),the time-frequency aggregation of Gaussian smooth filtered Hilbert spectrum is generally better under different SNR conditions.Finally,the K-MEANS algorithm based on amplitude-weighted and windowed is used to separate and extract the time-frequency lines of multi-component radar signals and estimate the parameters.Under the condition of 5d B,the proposed algorithm can effectively separate the mixed multi-component radar signals with four components.As well as the modulation parameter estimation,the parameter estimation error of the intra-pulse modulation of the radar signal of each component in the multi-component radar signal can be controlled within 5%.