Research On OTHR Track Initiation Algorithm

Due to its unique working mode,the sky wave over the horizon radar has the advantages of long range detection,stealth craft detection,low Detecting-Probability detection,and multi type target detection.The study of sky wave over the horizon radar is of great strategic significance.Data processing is an important part of the sky wave over the horizon radar system,and the track initiation is one of the key steps in data processing.in addition,there are few studies on the track initiation of the sky wave over the horizon radar at present,while the track initiation in the sky wave radar is faced with the problems of low detection probability,low data rate,low measurement accuracy and high false alarm rate.Therefore,the track initiation algorithm of sky wave over the horizon radar has the value to research.Aim at the characteristics of the sky wave over the horizon radar,we mainly study the two problems which influence the track initiation algorithm performance.The two problem we mainly study are the low detection probability and non-uniformly distribution of the clutter density.By comparing and analyzing the advantages and disadvantages of the existing track initiation algorithm and improving it,a special algorithm for the sky wave over the horizon radar is proposed,and the performance of the algorithm is verified by the processing of the simulated data and the measured data.Firstly,the target motion model and radar measurement model under the condition of sky wave over the horizon radar are established.According to the established radar measurement model,the conversion formula of the radar coordinate system and the geographic coordinate system is deduced.The correctness of the coordinate conversion formula is verified by the simulation experiment.Secondly,the advantages and disadvantages of the state estimation performance of Kalman filter and particle filter are compared and analyzed by simulation experiments under the established radar tracking model.The application of the RBPF filtering algorithm to the state estimation of the tracking model has been put forward.Then,track initiation algorithm under low detection probability environment and non-uniformly distributed clutter environment is studied respectively.We Simulate the logic method,the improved random Hough transform,the parallel flow based track initiation algorithm,the MHT track initiation algorithm in different detection probability environments,and propose an improved scoring function for MHT.A model of non-uniformly clutter distribution is established.Aiming at the model,a dynamic gate and a track recognition method based on comprehensive correlation score are proposed.Finally,the dynamic wave gate and the track confirmation method are applied to the improved MHT algorithm.Comparing the algorithm with the two level Hough transform algorithm,it is found that the algorithm has better initial performance.