Three-dimensional Passive Coherent Location Technology Considering Earth Curvature

Passive Coherent Location(PCL)system uses the third-party signals,such as FM radio,digital TV and mobile phone station,as the transmitter of opportunity.Through the signal coherent processing of the direct signal and the target echo signal,the target can be located and tracked.Due to the advantages of PCL in wide surveillance range,low cost,good concealment and strong anti-interference ability,the PCL system is of great significance to improve the detection performance of the military defense system.However,the elevation angel of the system is usually of low quality,which leads to the inaccurate height estimation.When the target is far from the receiver,the earth curvature further deteriorates the estimation accuracy.To solve this problem,a 3D passive coherent location technique taking into account the earth curvature is studied in this thesis.The main contributions are as follows:For the problem of 3D estimation with 2D measurements,such as range and bearing,in heavy clutter,a Probabilistic Data Association Height Parameterization(PDA-HP)method is proposed.Firstly,the probable height range of the target is divided into several height sub-intervals with equal weights.In each height sub-interval,the target state estimate is obtained using PDA.Then the weight of each height sub-interval is updated by the likelihood function,and the 3D target state is estimated by weighted fusion of all sub-interval states.Simulation results verifies the proposed method.When the target is far from the receiver,the estimate accuracy of the traditional method will decrease dramatically without the earth curvature.In this thesis,a method for the initialization and maintenance of 3D track considering the earth curvature is proposed.Firstly,the target state is established in the great circle coordinate system of the earth,the earth's great circle is transformed into the Earth-Centered Earth-Fixed(ECEF)coordinate system through coordinate transformation.The target dynamic model and measurement model including the earth curvature are established.Secondly,multi-frame measurements are used to construct the logarithmic likelihood ratio function,and the maximum value of this function is solved by a genetic algorithm,to initialize the target 3D track.Finally,the PDA is used to maintain the target track in clutter.The simulation results show that the proposed method can achieve better position accuracy than traditional methods.When the detection probability is low,the estimate accuracy and track continuity will decrease,a multistatic track-to-track fusion method is proposed for 3D PCL.Firstly,the target dynamic model including the earth curvature is established and complete target track initialization and maintenance to generate the local trajectory of each receiver.Secondly,the correlation terms of the cross-covariance matrix are derived by using the minimum upper bound criterion to calculate the cross-covariance between receiver approximately.Finally,the minimum variance fusion method is used to fuse the uploaded local track and obtain the final target state.Simulation results show that the proposed method can effectively improve the estimate accuracy and track continuity under low detection probability.