| With the intensification of the space competition among powerful countries in the world,more and more spacecraft launches are carried out in various countries,which leads to space debris flooding in low earth orbit and seriously threatens the safety of spacecraft and astronauts.The cataloging,surveillance and imaging of space debris have attracted wide attention of all countries.At the same time,the warhead in free flight phase and decoy target have the characteristics of high-speed rotation,precession and nutation.How to effectively monitor and identify the missile target has an urgent need in our country' missile early warning and air defense and anti-missile.Inverse Synthetic Aperture Radar(ISAR)has the advantages of all-day,all-weather,long-range and high resolution.ISAR is an effective means of space situational awareness.Because of the high-speed and non-stationary motion of space debris and ballistic target,the radar echo will have serious migration through range cells,and there are complex observation conditions such as missing echo data,occlusion of target and strong noise,which will make it difficult for conventional ISAR imaging algorithms to obtain well-focused imaging results.It is urgent to study new two-dimensional and three-dimensional high resolution imaging methods of rapidly spinning targets.Guided by the development trend of ISAR system and signal processing technology in the world,combined with the application requirements in the field of domestic space situational awareness,supported by the “13th Five-Year Plan” pre-research funds,“863” projects and lateral projects,the key issues that two-dimensional high-resolution imaging method,three-dimensional high-resolution imaging method and high-resolution imaging method based on sparse band fusion for rapidly spinning targets are studied in depth in this paper.The specific research contents of this paper can be divided into the following six parts:1.Firstly,the ISAR imaging turntable model and its basic imaging principle are introduced.Then,how to compensate the complex motion of the target to an equivalent ideal turntable model by translational compensation technology is expounded,and the range resolution and cross-range resolution of the image are analyzed.Then the difficulties of imaging high-speed rotating targets are discussed and the solutions are explained.Since the range resolution of ISAR imaging is inversely proportional to the radar bandwidth,how to realize coherent synthesis in the case of sparse frequency band is introduced in order to achieve the purpose of improving the imaging resolution,which lays a solid theoretical foundation for further research.2.Aiming at the problems that the radar echo of high-speed rotating target has large echo envelope migration and Doppler fast changes,a two-dimensional imaging algorithm of high-speed rotating target based on genetic algorithm is proposed.Firstly,the geometric model and signal model of high-speed rotating target imaging are established,and then the complex imaging problem is transformed into a parameterized sparse reconstruction problem.Then the envelope of each strong scattering point is extracted by improved genetic algorithm combined with CLEAN technology.The convergence speed of genetic algorithm is improved by adding appropriate genes to the population.Finally,the range-instantaneous Doppler images at different times are generated according to the parameters estimated.This algorithm can obtain good imaging results under complex observation conditions such as azimuth Doppler blurring or missing echo data,and its operation efficiency is better than the existing parameterized ISAR imaging algorithm of high-speed rotating targets.3.In order to improve the classification and recognition ability of high-speed rotating targets such as space debris and ballistic targets,it is necessary to acquire more abundant structural information than two-dimensional images.The echo of high-speed rotating target has serious migration through range cells,and the radar pulse repetition frequency is low because the space debris and other targets are far from the ground.Under this condition,the echo will have serious Doppler ambiguity,which makes the traditional three-dimensional imaging algorithm difficult to deal with.In order to solve the above problems,this paper presents a three-dimensional imaging algorithm for high-speed rotating targets based on Inverse Radon Transform(IRT)and CLEAN technology.Firstly,the principle of three-dimensional imaging of high-speed rotating target is deduced,and then the complex imaging problem is transformed into a parametric constrained optimization problem.According to the improved particle swarm optimization(MPSO)algorithm,Z coordinates and trajectories of strong scattering points are extracted,and then the improved CLEAN technology is proposed to extract the envelope of strong scattering points one by one.The output of MPSO algorithm is used as the input of IRT to estimate the X and Y coordinates of scattering points.Finally,the coordinates of scattering points are calibrated to obtain three-dimensional images.Compared with the traditional three-dimensional imaging algorithm,the proposed algorithm still achieves good imaging results under the conditions of self-occlusion,data missing and azimuth Doppler ambiguity.4.In order to overcome the difficulties in estimating the number of scattering centers and the large errors in calculating the rotation period of high-speed rotating targets in the three-dimensional imaging algorithm based on IRT,a three-dimensional imaging algorithm for high-speed rotating targets based on non-parametric Bayesian model is proposed.The MPSO algorithm combined with M-CLEAN technology is used to calculate the height of each scattering point,and then a non-parametric Bayesian model is established to accurately estimate the target rotation period,range direction,azimuth coordinates and relative RCS intensity.In particular,the algorithm uses reversible jump Markov chain Monte Carlo algorithm to estimate the complex posterior probability distribution of scattering points.Compared with the existing high-speed rotating target three-dimensional imaging algorithm,the proposed algorithm has higher reconstruction accuracy and good robustness to low signal-to-noise ratio,azimuth Doppler ambiguity and data missing.5.Aiming at the problems of simple model and limited precision of scattering center reconstruction in traditional sparse band fusion technology,an ISAR sparse band fusion model based on target attributed scattering center model is established in this paper.Then the source of coherent phase between sub-bands is deduced,and then the scattering center parameters of different sub-bands are extracted according to the scaled dictionary technology.Compared with the traditional method,the time complexity of this method is lower.Then the sparse representation and particle swarm optimization algorithm are used to calculate the coherent phase.Finally,the scattering center of the attributes of the synthesized full-band signal is extracted to further reduce the calculation error.6.At present,the bandwidth of a single radar is limited.How to fuse sparse band signals using multiple radar working in different frequency bands to improve the observation ability of ISAR equipment for high-speed rotating targets deserves further study.In order to solve the problem that the sparse frequency band signals of high-speed rotating targets are difficult to fuse when different frequency bands of radar are far distributed,a matrix beam algorithm is proposed to fuse radar signals of different frequency bands.Based on the fused wide bandwidth signal,the non-parametric Bayesian model is used to solve the three-dimensional imaging results of high-speed rotating targets.Compared with the traditional single-band three-dimensional ISAR imaging algorithm,the proposed algorithm can obtain three-dimensional images of high-speed rotating targets in sparse frequency band under the condition that different radar bands have different observation angles. |
PDF Full Text Request