Research On Long-Time Coherent Processing Algorithms For Near-Space High-Speed Target

Near-space high-speed targets show the characteristics of supersonic,low visibility,high maneuverability and strong penetration ability,imposing grim challenge to the radar early warning detection.With the improvement and advancement of modern radar technologies such as array radar,beamforming and digital signal processing,modern radar can increase the echo signal-to-noise ratio thereby improves the detection performance through the long-time staring pattern,which could be realized by coherently processing multipulse echo signals.However,as the increase of observation time,the target motion components(velocity,acceleration and jerk)will cause range migration(RM)and Doppler frequency migration(DFM)of the echo signal,affecting the coherent processing performance.Regarding the problems of RM and DFM for the near-space high-speed targets with different motion modes(velocity,acceleration and jerk motion),this dissertation has carried out theoretical analysis,algorithm research,simulation verification and other work,and the main innovations are given as follows:1.The coherent processing algorithm for the target with constant velocity based on modified location rotation transform(MLRT)is proposed.The MLRT corrects the first-order range migration(FRM)by rotating the coordinate location of echo peak,and then realizes the signal energy coherent integration via slow time Fourier transform(FT),which can improve the signal-to-noise ratio gain.2.The coherent processing algorithm for the target with constant acceleration based on Keystone transform-matched filtering process(KT-MFP)is proposed.The KT-MFP firstly compensates the FRM caused by target's unambiguous velocity,and then eliminates the residual RM and first-order Doppler frequency migration(FDFM)by two-dimentional MFP.The KT-MFP can avoid the blind speed sidelobe(BSSL)effect and achieve good coherent integration results.3.The coherent processing algorithm for the target with constant jerk based on bilinear parametric symmetric self-correlation transform-scaled Fourier transform-improved second-order phase differentiation(BPSST-SCFT-ISPD)is proposed.The third-order range migration(TRM)is corrected by the third-order KT(TKT),and then the target's velocity,acceleration and jerk are estimated by BPSST,SCFT and ISPD.Next,constructing the phase compensation function to eliminate the residual RM and DFM.The BPSST-SCFT-ISPD can avoid multi-dimensional parameter searching and balance the relationship between computational complexity and integration performance.4.The scale echo model of ultra-high-speed targets under the large time-bandwidth product is established,and the coherent processing algorithm based on scaled Radon Fourier transform(SCRFT)is proposed.The SCRFT realizes the matched compression of intra-pulse signal and the inter-pulses coherent integration simultaneously,which improves the coherent processing performance.The above algorithms have been verified by theoretical analysis and simulation experiments.The results indicate these algorithms could availably address the problems during the long-time coherent processing and realize the coherent integration of energy.