Research On Refined Feature Extraction Method Of Warhead Target

Compared to early narrowband radars,broadband radars have high range resolution,strong anti-interference ability,and good electromagnetic compatibility.Their electromagnetic echoes contain more abundant electromagnetic spectrum information.Therefore,more effective and accurate refinement features can be obtained for warhead type targets,which can distinguish warheads from objects such as bait and debris.Currently,the main methods for extracting refined features of targets using wideband radar echoes are super resolution methods and Fast Fourier Transform(FFT)methods.However,the former usually has inaccurate estimation of model order,while the latter has low accuracy in test results.Therefore,a comprehensive use of the two methods can be considered to achieve complementarity.Currently,research on warhead targets usually assumes that they are stationary or translational,and the effects of high-speed motion are also ignored when considering the effects of their micromotion.Therefore,this thesis comprehensively considers the simultaneous high-speed motion and micromotion of warheads.And conducts research on methods for extracting refined features of warheads in motion.The specific work is as follows:Firstly,a target micromotion model is established for a warhead target with micro motion,which is used to describe the relative distance change of the scattering center on the target.Then,an echo signal model is established based on Geometric Theory of Diffraction(GTD).And based on the target High Range Resolution Profile(HRRP),the change in the test length of the target,and the change in the cosine value of the angle between the target’s radial direction and the radar line of sight are observed.Next,aiming at the method of extracting target length from HRRP in experiments,a more efficient scattering center extraction method based on priority queues is proposed compared to FFT algorithm and adaptive algorithm.Then,for high-speed moving targets in the actual scene,the impact of target speed on the obtained HRRP is analyzed,and the impact of speed is compensated using the entropy minimum criterion method to improve the distortion phenomenon,so as to obtain the radial length characteristics of the target more accurately.Finally,experiments are conducted to verify the impact of high-speed motion and micromotion on the target echo signal,as well as the changes in attitude angle and test length.Due to the error impact caused by the range resolution of the target HRRP obtained in practice,the test length curve is not smooth.Therefore,this thesis proposes a secondorder sinusoidal fitting method based on the least square method.The results show that this method can improve the accuracy of target length estimation.Then,a traversal search method is proposed to find the initial attitude angle of the target within a micromotion cycle,which is used to estimate the true length of the target.It is verified that the estimated target length obtained is closer to the actual value when considering the simultaneous high-speed motion of warhead targets than when considering only micro motion.Finally,FEKO is used to simulate three types of warhead targets to verify the method in this thesis.The results show that the method proposed in this thesis is more accurate compared to previous methods that did not consider the high-speed movement of warhead targets,or only used one structural feature extraction method.