| Near-space vehicles have more and more extensive applications in military fields such as rapid defense penetration,reconnaissance surveillance and communication navigation due to their hypersonic speed,strong maneuverability and low observability.However,the hypersonic and strong maneuvering characteristics of such targets will cause the echo signal to produce range migration and Doppler frequency migration,resulting in the traditional coherent integration method will no longer be applicable.Therefore,it is an urgent need to study the accumulation detection method suitable for near-space hypersonic targets in the current radar field.This paper focuses on this problem in terms of basic theory and algorithm simulation.The specific contents are as follows:1.The basic theory and algorithm of hypersonic target detection in the near space are studied.Firstly,the echo model of the high-speed target in the near space is established,and the range migration and Doppler frequency migration are analyzed theoretically and experimentally.Then,for the problem of range migration,three correction algorithms,namely Keystone transform method,Radon-Fourier transform method and frequency domain phase compensation method,are studied.For the Doppler frequency migration problem,two compensation algorithms,Dechirp method and FRFT method,are studied.Finally,the merits and demerits of each method are analyzed and summarized through simulation experiments.2.The hypersonic multi-target detection algorithm in the near space is studied.Firstly,the echo model of multiple high-speed moving targets is constructed.For the situation that the target echo intensity is similar,two algorithms,KT-CICPF and KT-MFP,are studied.The KT-CICPF algorithm corrects the range migration and Doppler frequency migration in turn through KT processing,ambiguity number search and CICPF parameter estimation to achieve coherent integration of energy,and the algorithm uses CICPF to effectively eliminate the interference of cross terms.The KT-MFP algorithm realizes parameter estimation and energy accumulation by constructing a two-dimensional matched filter function,and effectively solves the problem of range curvature.Then the simulation experiments prove that both algorithms can achieve multi-target accumulation detection in different multi-target scenarios.In order to solve the multi-target detection problem of coexistence of strong and weak targets,two coherent integration algorithms based on CLEAN processing are studied.Under the condition of obvious difference in target scattering intensity,both processing methods can achieve multi-target coherent integration,and the improved CLEAN algorithm has a higher tolerance for target parameter estimation errors.3.Aiming at the problem that the coherent integration algorithm based on velocity search and Dechirp processing requires a large amount of computation,a segmentation-based hypersonic target detection algorithm in the near space is proposed.The proposed algorithm adopts a combination of coherent integration and non-coherent integration.Firstly,the echo signal is segmented along the slow time,and then the intra-segment coherent integration is achieved through velocity search and Dechirp processing in the sub-segment,and finally the inter-segment non-coherent integration is completed according to the velocity and acceleration search values.The performance analysis and simulation results of the algorithm show that the proposed method is able to gain good balance between the amount of computation and accumulated performance,which is easily applied to engineering. |
