Research Of Small Uav Target Detection Based On Long-time Observation

Radar detection is an effective way to find UAV target.However,small UAV target has the characteristics of small size,variable motion state,and complex flying environment,which make the signal-to-noise ratio(SNR)of the echo signal weak and the radar detection full of challenges.The long time integration method can increase the echo signal-to-noise ratio(SNR)by integrating more echo signals.However,the target echoes often accompany range migration(RM)and Doppler frequency migration(DFM)problems during long time integration,which make the effect of traditional coherent integration method decrease significantly.Therefore,correcting RM and DFM of UAV target is key important to improve radar detection performance.The motion state of small UAV target is complex.In the radial direction,the small UAV target not only has velocity but also has acceleration or higher-order motion parameters during flight.This thesis establishes three different echo models of small UAV target and studies the long-time integration algorithms of different echo models under strong clutter background.The following are the specific content of the thesis:Aiming at the RM caused by moving UAV target with radial velocity.In this thesis,two typical long-time integration algorithms of KTP and RFT are studied to correct RM of the target echos.Simulation experiments and data processing of measurements confirm that both algorithms achieve satisfying detection performance to small UAV target in strong clutter background.The detection performance of KTP algorithm is better in low SNR environment,while the computational cost of RFT algorithm is lower.Considering the detection performance and computational complexity,when Doppler ambiguity is considered,RFT algorithm is more suitable;when Doppler ambiguity does not need to be considered,KTP algorithm can be selected.The moving UAV target with radial acceleration not only has RM,but also has DFM.Aiming at these problems,KTPDP and KTPLVD algorithms are studied.Simulation experiments and data processing of measurements confirm that both algorithms achieve satisfying detection performance to small UAV target with acceleration in strong clutter background.The detection performance and computational complexity of the two algorithms are basically the same when detecting low-speed flying UAV target.Through overall consideration,both of them are suitable for rapid detection of UAV target with low-speed in low SNR environment.Aiming at the moving UAV target with radial jerk,compensation for Doppler curvature is also required.In this thesis,two long-time integration algorithms of KTPGDP and ACCF-LVD are studied.Simulation experiments and data processing of measurements confirm that both algorithms achieve satisfying detection performance to small UAV target with jerk in strong clutter background.The detection performance of KTPGDP algorithm is better in low SNR environment,while the computational cost of ACCF-LVD algorithm is lower.Considering comprehensively the detection performance and computational complexity,although ACCF-LVD algorithm ows faster detection speed,the effect to resist noise is relatively weak.KTPGDP algorithm is more suitable for detecting UAV target in lower SNR environment.