Research On Detection Algorithm Of Radar Low-altitude Slow And Small Targets

Low-altitude slow and small targets such as UAVs and stealth aircraft are difficult to be detected by traditional low-resolution radar due to their low flight height,slow flight speed and small radar cross-sectional area(RCS).Therefore,the detection of such targets has always been a hot research issue.This paper focuses on the radar low-altitude slow and small target detection algorithm and related tracking problems.The specific work is as follows:1.The characteristics of radar clutter and the basic theory of target detection are studied.The characteristics of conventional radar clutter and the modeling method of clutter echo are introduced to prepare for the research of clutter suppression and target detection algorithm.The prior conditions of fixed threshold detection and constant false alarm rate(CFAR)detection algorithms are compared,and the detection performance of the two algorithms is simulated and analyzed.The similarities and differences of the algorithm implementation process of detection-before-tracking(DBT)and track-before-detection(TBD)techniques are analyzed.2.Research on zero-frequency clutter suppression algorithm for ground clutter environment.The time domain model of radar clutter is established,the moving target indication(MTI),moving target detection(MTD)and Kalmus filter are designed,and the frequency response results of each filter are analyzed.The signal-clutter-noise ratio(SCNR)improvement capability of each filter’s low-speed channel under different clutter intensities is verified through performance simulation and measured data.3.The slow target detection algorithm based on clutter map is studied under the framework of DBT technology.Based on the background of Rayleigh distribution clutter,the realization process and related parameter design of clutter map CFAR point technology and plane technology is studied,and the influence of different parameters on detection performance is compared.Based on the measured data,the detection capabilities of various mean-level CFAR detection algorithms and clutter map CFAR detection algorithms for slow targets are compared.The results show that the clutter map established by the MTD low-speed channel and Kalmus-filtered data has better detection ability of slow or near-zero-speed targets because it both suppresses the zero-frequency clutter and preserves the slow target information.4.The maneuvering weak target detection algorithm based on PF-TBD is studied under the framework of TBD technology.The radar TBD model is established,and the detection and tracking process of the PF-TBD algorithm is theoretically deduced and simulated in detail on the basis of this model,and the detection and tracking performance of the algorithm under different signal-to-noise ratios and different particle numbers are analyzed.A multi-mode PF-TBD algorithm based on the hybrid estimation theory is studied for maneuvering weak targets,and the detection and tracking performance loss caused by the multi-mode PF-TBD algorithm is improved for the weak target motion state mutation scene.The experimental results of simulation and measured data show that the improved algorithm has better detection performance and tracking error.It also shows that the PF-TBD algorithm has better low-altitude slow and small target detection ability than the traditional single-frame target detection algorithm.