Research On Intrabeam Group Target Parameter Estimation And Tracking Technology

With the development of modern radar detection technology and the increase of the target and environment complexity,there often exists the problem of motion state estimation of group target.The resolution of the radar determines whether reliable and stable parameter estimation and tracking of the target can be achieved.Since early tracking radars mainly use the envelope amplitude information of one-dimensional range profile for range and velocity estimation,and the target motion estimation and tracking process are relatively independent,they cannot distinguish adjacent ones of the group target within one antenna beam.This kind of processing has great problems in the group target tracking scene,and is difficult to make a substantial improvement in signal processing limited by the signal model.Therefore,in order to solve the problem of parameter estimation in group target scene,we can separate the group targets by adding velocity observation dimension in signal processing and track the group targets through filtering and data association,and finally use tracks to assist signal processing to separate individual target and resolve the velocity ambiguity.Thus,the following aspects of research are expanded in this paper.Firstly,the signal model of narrow band LFM radar is studied in detail,and the digital signal processing in practical radar system is described as well.A simple ballistic model is built to generate the simulated echo signal,which provides the signal basis for parameter estimation.Traditional methods of parameter estimation are mostly based on the complex envelope information of one-dimensional range profiles,which are not able to be applied in group target scene.Therefore,we use the method of maximum likelihood estimation.This method can achieve the theoretical optimal estimation precision,but it also has great computational complexity.Based on IFFT,CZT and data compression,three fast implementation methods are proposed to reduce the complexity.Due to the distribution of the likelihood function,there may exist estimation deviations caused by the superposition of range-velocity cross side-lobes,and the estimation of velocity could be ambiguous.Limited by the signal characteristics,these problems cannot be solved simply by signal processing.In order to solve the problems for MLE algorithm,data processing which includes the filtering and data association is applied after parameter estimation.Firstly,a simple comparison between standard Kalman filter and adaptive Kalman filter is carried out,and then through the data association technology,the individual stable tracks are separated in group target so as to solve the velocity ambiguity problem.Thus the independent frameworks of parameter estimation and target tracking are combined to form a close-loop system,and the parameters estimation of the group target can be achieved in high precision.