Research On Improving The Accuracy Of Moving Multi-Station TDOA Passive Location

In modern warfare,the reconnaissance and positioning of enemy radar plays a key role in the success or failure of the war.Passive location has the natural stealth characteristics,and can locate the target without the enemy's knowledge.Multi-station passive location based on time difference of arrival of signals is one of the important research contents of passive location because of its simple structure and high positioning accuracy.UAV fleet is more and more widely used in modern warfare,especially in target reconnaissance.Because UAV can maneuver flexibly and keep working together,the multi-station TDOA positioning system based on UAV fleet can approach the target radar and fly in formation maneuverably according to the positioning needs to achieve high-precision passive positioning of the target.In this paper,aiming at improving the positioning accuracy of the passive location system with moving multi-station TDOA,three main factors affecting the positioning accuracy are studied,which are positioning algorithm,signal arrival time difference measurement error and station location measurement error,and the methods of improving positioning accuracy are discussed.For multi-UAV TDOA positioning system,the key problem is to solve the non-linear equations of target position.The main methods can be divided into three categories: analytical method,iterative method and search method.The positioning error of analytic method is large,and Newton iteration algorithm is widely used in engineering because of its simple algorithm,easy implementation and stable performance.Compared with traditional Newton iteration algorithm,classical particle swarm optimization algorithm can obtain higher positioning accuracy.Similar to particle swarm optimization,artificial bee swarm optimization algorithm is also a search algorithm for population optimization.Therefore,this paper studies it.Its application in TDOA positioning system is studied.Because TDOA positioning system optimization problem is a classical quadratic constrained quadratic optimization problem,because the constraints are not convex sets,it can not be solved directly by convex optimization theory.In this paper,the problem is transformed into Semidefinite Programming(SDP)problem after deforming the positioning equation and relaxing the conditions.A convex optimization problem can be solved by Lagrange multiplier algorithm.By comparing Root Mean Square Error(RMSE)and Geometric Dilution of Precision(GDOP)of several algorithms,it is found that the positioning accuracy of artificial bee colony algorithm and Newton iteration algorithm is comparable.The positioning accuracy of particle swarm optimization algorithm is better than the former two,and the positioning accuracy of semi-definite programming algorithm is the highest.Therefore,to improve the positioning accuracy of the system,the semi-definite relaxation programming algorithm is applied to solve the positioning equation.Then,in order to reduce the influence of measurement error of time difference of arrival on positioning accuracy,this paper studies the precise time difference extraction algorithm of correlation peak interpolation,which is a precise method of time difference of arrival.It mainly discusses several methods of estimation of time difference of arrival of signal: correlation function method,generalized correlation function method and correlation peak interpolation method.The research shows that under the same sampling method,the error of arrival time difference measured by this method is only about one fifth of that of the traditional electronic counting method,and the maximum can be less than 1ns,which can significantly improve the positioning accuracy of the system.Finally,another method to improve the positioning accuracy of the system is to increase the observation information of the reference emitter whose position is known.The observation information of the reference emitter is substituted into the solution equations of the target position to offset the measurement error of the station location,so as to improve the positioning accuracy.This paper discusses the situation of virtual reference sources under the condition of single reference source and multi-reference source respectively,deduces the equation of target location under the condition of adding observation information of reference source,and then applies Newton iteration algorithm to solve the new equation of target location,and compares the result with that of traditional Newton iteration algorithm.The results show that the positioning accuracy of the system can also be effectively improved by increasing the observation information of the reference source.