Research On Passive Positioning Technology Based On UAV Platform

In modern battlefield,electronic reconnaissance plays an important role in electronic warfare.Passive positioning systems have good concealment characteristics and are not easily detected by enemies when positioning the radiation source,which is a crucial function of electronic reconnaissance.As unmanned aerial vehicle(UAV)become the new favorite for electronic reconnaissance,particularly in target reconnaissance,their applications are becoming increasingly widespread.Therefore,considering the high mobility of UAVs,this paper focuses on methods to improve target localization accuracy in the presence of station error.The study is conducted based on the theoretical basis of time difference localization,factors influencing localization accuracy,time difference localization algorithms,and the application of neural network in time difference localization.Firstly,the paper explores the principle of target localization using time difference,including theories and techniques such as measurement and time synchronization of time difference.For target localization using multi-UAV cooperation,the paper examines commonly used coordinate systems and the transformation relationships between them,including providing the equation for the transformation relationship between the geodesic coordinate system and the station-centered coordinate system.Secondly,the paper establishes a time difference positioning model in three-dimensional space and conducts a detailed study on the factors influencing the positioning accuracy in the time-difference positioning system.The study analyzes the effects of the number of UAVs,baseline length,station site error,time difference measurement error,target source location,and UAV formation structure on the target localization performance through simulation experiment.Then,building on the time difference positioning model,the paper derives the Cramer-Rao Lower Bound and geometric accuracy factor.It summarizes the classical localization algorithms,including Chan and Taylor algorithm.In the presence of UAV station site error,the paper improves the above algorithms and proposes an improved Chan-Taylor weighted joint algorithm.The simulation verifies that the algorithm can reach the Cramer-Rao Lower Bound when the station error is small,and it performs well in terms of positioning accuracy when the station error is large.Finally,the paper describes a multi-UAV cooperative target localization scenario for locating radiation source targets in a specific region.It explores the application of the back propagation(BP)neural network in time difference localization in the presence of station error.To avoid the impact of very sensitive initial network weights and slow convergence of BP neural network on localization performance,the paper proposes a genetic algorithm to optimize the localization algorithm of BP neural network.The feasibility and effectiveness of the algorithm to achieve target localization are analyzed by simulation,and it is verified that the station error has a negligible effect on the root mean square error of the algorithm,which can almost reach the Cramer-Rao Lower Bound.The algorithm also shows strong robustness and short localization time in the scenario of multiple UAVs and multiple targets.