Reaserch On Track-before-detect Algorithms Combining With Real Data

The detection and tracking of radar weak targets has always been an important issue.Traditional Detect Before Track(DBT)technology cannot reliably detect targets in a complex environment with low Signal Noise Ratio,and Track Before Detect(TBD)technology can multiple scanning methods are used to increase the Signal Noise Ratio in the case of strong clutter,strong interference,and low Signal Noise Ratio,thereby improving detection and tracking performance of radar weak targets.In this thesis,based on the relevant theories at home and abroad,the research and analysis of Particle Filter TBD for radar weak targets are carried out.The research contents are as follows:1.For the basic radar target detection and tracking technology,through the analysis of the traditional DBT technology and the current rapid development of TBD technology,and then summed up the advantages and disadvantages of the two technologies.2.Start with the basic concepts of nonlinear filtering theory,emphatically introduce the basic principles and algorithm procedure of the basic particle filter algorithm(PF).Through a classical system model,use several methods,such as Extended Kalman filter(EKF),Unscented Kalman Filtering(UKF)and particle filter(PF)are compared.It is concluded that the PF method has a better estimation effect for nonlinear systems,and under the same conditions,increasing the number of particles applied in the PF algorithm can enhance the tracking effect in a nonlinear system.3.First introduce the TB-based TBD algorithm,describe the general procedure of radar signal processing for TBD.Secondly,use a single-target uniform linear motion in a two-dimensional plane as the target motion model.and the corresponding TBD processing model is established in the radar application for the TBD technology.Finally,through the simulation analysis and comparison of the target detection probability and track tracking performance,and based on the problems of the basic algorithm,several improved forms of PF-TBD algorithm are expounded,and the advantages of several improved form algorithms are illustrated by simulation experiments.The simulation analysis shows that a reasonable selection of the particle distribution unit set of the improved algorithm can improve the performance of detection and tracking compared with the basic PF algorithm.4.Based on the urgent requirements of the current TBD technology in engineering applications,combined with the radar echo data collected from the external test site,according to the PF algorithm introduced in this thesis,the contrast experiments of different signal to noise ratio and different signal to clutter plus noise ratio are carried out with the traditional DBT method,The comparison experiment below summarizes the advantages and disadvantages of traditional DBT method and Particle Filter Based TBD method.Simulation experiment shows that Particle Filter Based TBD method has better results,and in low signal to noise ratio and low signal to clutter plus noise ratio.The advantage is even more obvious.