Research On Bearings-Only Single Station Passive Location And Tracking

Bearing only positioning and tracking has been a very important way in passive system, which mainly abstracts the angle information from the target emitted radiowave received by passive sensors array. Then the angle information will be malipulated to localize and track target through certain fomula between target and monitoring station, thus high accuracy of angle shall be needed. Because of the monitoring station`s continuous movement along with the uncertain moving state of target, the observability analysis of moving object is necessary. The main work of the thesis is as follow:Firstly,the theory of positioning and tracking of moving target based on bearing only mothod is introduced and the observability is analyzed according to different moving state of target.Secondly,multiple target localizing method base on subspace data fusion is proposed, which mainly analyzed the multiple motionless target positioning using monitoring station making uniform linear motion, samples on different time spot and finally puts the bearing matrix sampled at different time spots together which can finally be used to get the position of each target through subspace data fusion algorithm where setting up the cost function of bearing and target coordinates` linearity and peak scanning will be used. This method only calls for one cost function to get the optimized estimator instead of DOA estimation of traditional method, thus greatly reducing the calculating amount and resulting more locationing accuracy.Thirdly,the traditional tracking filtering algorithm based on bearing only method and the tracking module are introduced. Modified Gain Extended Kalman Filtering(MGEKF), Modified Virance Extended Kalman Filtering(MVEKF),Unsensitive Kalman Filtering(UKF) algorithms are mainly introduced, which do the unlinear transform of the few samples regardless of linear processing, thus achieving better accuracy. Then the simulation is done for these filtering algorithm and corresponding performance on different simulation scene is analyzed.At last, the tracking filtering algorithm using data fusion based on optimum weighting method is raised to improve the performance of traditional flitering algoritm. Partial observations for each monitoring station is used to get the partial estimation using UKF filtering algorithm, then data fusion based on optimum weighting method is used to process these estimation then to get the overall filtering estimation. This method will improve the filtering accuracy and finally the performance comparison and analysis is made between this proposed method and traditional filtering algorithm.