Feature Extraction And Recongnition Of Aerial Targets In Frequency Agile Meter Wave Radar

Because of its excellent anti-jamming ability and anti-stealth ability,the frequency agile meter wave radar has received extensive attention from domestic and foreign scholars and has been widely used in the military field.For general aerial targets,a large number of frequency point data in the target resonance area can be obtained at one time through the frequency agile meter wave radar.Many features of the target can be obtained through these data,and the pole feature is the most stable feature among the many features,because the pole feature can remain unchanged under different incident wave directions and polarization modes.It is related to the size,material and shape of the target itself.The pole feature can be used for the recognition of radar targets,used to distinguish the enemy,distinguish the target type and the carrying situation of the missile,so this article mainly studies how to use the pole feature to target different types of air targets in the context of frequency agile target radar.Identify it.This article first gives a brief introduction to the related theories of the frequency agile meter wave radar,and analyzes the characteristics of the radar cross-sectional area(RCS)in detail;analyzes the relationship between the RCS and the target pole from the singular point expansion method and the system function.It also explains the solution process of radar target RCS.Then,two pole extraction methods,Cauchy and MPM,were used to extract the pole features of simple targets and complex aerial targets.The extraction effects of the two algorithms were analyzed and compared;the two methods under different signal-to-noise ratios were also analyzed.The pole extraction effect of the method;the poles extracted from the measured data are used,and the poles extracted from the simulation data are compared and analyzed.Then this article considers the missing of some frequency point data,because in the context of actual frequency agile meter wave radar,the echo data is likely to be unavailable or missing at some frequency points due to noise and interference,so this problem needs to be addressed.solve.This paper uses polynomial interpolation and compressed sensing to recover the missing frequency point data of different targets,and extracts the extreme point features of the recovered data,and analyzes the effect of the extraction.Finally,the recognition of aerial targets under the background of frequency agile meter wave radar is studied;that is,based on the pole features extracted above,three different types of aerial targets are identified and the recognition results are analyzed with support vector machines.