Research On Radar Target Identification In The Near Space Hypersonic Condition

Near space hypersonic aircraft is a hot spot in recent years,which is usually considered as an upper bound between height in general aviation aircraft and aircraft flying in the lower airspace.Aircraft which is faster than Mach 5 is usually considered as a hypersonic aircraft.Comparing with traditional types of air vehicle,hypersonic aircraft has less radar scattering cross section for their design in stealth.The high speed of hypersonic fight lead to some problems just like echo signal walking and phase ambiguity which will weaken the performance of radar detection and recognition seriously.Carrying out researches on motion of hypersonic aircrafts and radar electromagnetic characteristics affected by aerodynamics shape in high velocity which will be used in radar recognition is very important.This paper studies the theories?models and fusion identification methods of hypersonic target recognition in radar,including:Firstly,this paper introduces the principles and mechanisms of radar target recognition,studies the application scope and availabilities of different radar recognition technologies.It is necessary to point out difficulties and features of target recognition and ways extraction combined with the existing design of hypersonic vehicle testing program.Secondly,based on the characteristics of high speed target and radar scattering characteristics,the methods of radar identification are presented.Remote narrowband radar is used to get range-doppler data which described features such as altitude?speed and some other "rough" things of targets to be identified.And then,bispectrum in one dimensional profile extraction?parameterized model of scattering centers and range-instantaneous Doppler features in ISAR were studied in target image recognition in broadband radar.Finally,the paper studies multi-sensor fusion target recognition and the fuzzy theory with its application to solve uncertainty problems in fusion recognition.Mechanisms in each layers of fusion recognition information have been studied,decision fusion recognition available in heterogeneous multi-sensor fusion recognition is applied for hypersonic vehicles identification.The confusion matrix,genetic algorithm and adaptive fuzzy density value assignment method are analyzed for fuzzy density,including their applications and disadvantages.From simulation calculations in training samples which obey Gaussian or uniform distribution or their all,those three methods are compared in the fusion recognition average rate computed through fuzzy density determined by each method.