Research On Micro-motion Feature Analysis And Extraction Technology Of Ballistic Target

As a key technology in ballistic missile defense system,the feature extraction and classification of trajectory micro-moving target is a problem that must be solved in modern warfare.Based on radar automatic identification application of the micro-moving target,this thesis follows the logical structure of target modeling-radar characteristic analysis-feature extraction-classification recognition.The feature extraction of single / multi-component sinusoidal frequency modulation signal is used as the main clue to study the feature extraction technology and parameter estimation and other issues of micro-moving target.The main work and research results of this dissertation include:1)The trajectory target micro-motion model and the all-polar scattering model are established.First,based on the micro-motion model,the closed expression of the target geometry,the micro-moving parameter and the echo micro-doppler frequency is given,and a unified representation of the micro-doppler frequency of the basic micro-moving type is given,and the time-frequency characteristics are analyzed.Secondly,the polarimetric scattering geometrical model is established,and the transformation relation between the radar emitter base and the target polarized base is deduced.The two models are connected with the attitude angle and the polarization rotation angle.2)A method of synthesizing the full-polarized radar echo and polarization deception jamming signal is proposed.Both methods can accurately simulate the micro-motion characteristics,structural characteristics and polarization characteristics of the target.First,the given echo synthesis process can simulate a specific micro-moving target,and can produce a variety of micro-motion characteristics of the polarization radar waveform;Second,the traditional DIS method that based on the fixed image template is improved;the proposed polarization deceptive jamming signal synthesis method,can dynamically generate false target modulation coefficient,which has high fidelity and good real-time.Both can provide data support for feature extraction and classification recognition.3)A MLE method based on RCS feature is given to estimate the parameters of single component sinusoidal frequency modulation signal.The RCS fluctuation characteristics of the trajectory micro-moving target are briefly introduced,and the MLE of the FM parameter under the constant and complex Gaussian amplitude is compared and the CRLB is deduced.At the same signal to noise ratio,the latter's estimated performance is higher than the former,indicating the use of RCS features to estimate the FM parameters need to consider RCS fluctuation;4)The reassigned time-frequency-hough transform is proposed to estimate the multi-component sine-frequency modulation phase signal parameters.The reassigned time-frequency algorithm not only preserves the cross item suppression performance and good anti-noise performance of the bilinear time-frequency analysis method,but also improves the energy clustering property and can improve the accuracy of the time-frequency curve of the Hough transform.5)A hierarchical clustering method based on micro-Doppler time-frequency curve is proposed to identify active deception jamming targets.By using the spatial diversity characteristics of the networked radar,the time-frequency curve can be extracted directly by the clustering method.6)The radar identification system of ballistic active deceptive micro-moving target is designed.The simulation system of active deceptive target radar is given,and the system simulation is carried out to verify the validity of the proposed model and algorithm.