| The outcome of modern warfare depends on the strength of both sides in electronic warfare.In electronic warfare,passive reconnaissance positioning and tracking technology are main methods that are very important and widely used.After many years of development,several mature passive TDOA positioning algorithms and tracking algorithms have been formed,but the accuracy of them needs to be improved urgently,especially in the case of high time difference measurement errors.In recent years,with the gradual rise of meta-heuristic algorithms,traditional TDOA positioning algorithms and tracking algorithms have developed a new direction.Among them,particle swarm optimization(PSO)algorithm has been widely used in passive TDOA positioning algorithms and tracking filtering algorithms,and have higher accuracy.But it has shortcomings such as falling into local optimum easily and so on.Based on the characteristics of different meta-heuristic algorithms,this thesis selects the recently presented salp swarm algorithm(SSA)with a stronger ability to jump out of the local optimality,which has strong individual diversity and high accuracy,and applies it to passive TDOA positioning algorithms and filtering algorithms.Improve separately on the SSA that is applied to position and tracking.An improved salp swarm algorithm(WSSA)and a particle filter based on salp swarm algorithm(SSA-PF)are presented.They get better performance.The research content of this thesis is introduced as follows:Firstly,the basic principle of TDOA positioning and the classic algorithm such as Chan algorithm and Taylor algorithm are researched.TDOA positioning based on PSO algorithm is researched.Several indexes for measuring positioning performance are introduced.This thesis mainly uses the RMSE to measure the performance of the algorithm.The positioning accuracy of the classical algorithms and PSO algorithm are compared by simulation,and the advantages and disadvantages are analyzed.Secondly,the classical filtering algorithms and motion models in single target tracking are researched.The performance of EKF algorithm and UKF algorithm in the univariate nonstatic growth model is analyzed by the simulation.Based on UKF algorithm,the performance of the Singer model is compared with the IMM for the target with a complex motion path.Then,SSA and its inspiration source of biological are introduced.SSA is applied to TDOA positioning.SSA with low accuracy and weak search capabilities is improved,enhancing the search ability through increasing the number of leaders and the ability to jump out of the local optimum through Gauss disturbance.WSSA is presented.The individual distribution of the population in the early stages of algorithms shows that WSSA has a stronger ability to jump out of the local optimum to obtain higher accuracy than SSA.The stability in different spatial positions,the convergence rate and convergence accuracy of the PSO algorithm and the WSSA on TDOA positioning are compared by the simulation,and the advantages and disadvantages are analyzed.Finally,a standard particle filtering(PF)algorithm was derived from Bayesian filtering,and the SSA was used to optimize the particle filtering.A particle filter based on salp swarm algorithm(SSA-PF)was presented.The particle movement is guided by SSA,so that the particles are mainly distributed in the area around the true value.The problem of particle weight degradation is improved and the accuracy is improved.At the same time,there are particles in other areas.The SSA-PF is compared with the particle filter based on particle swarm optimization(PSO-PF)algorithm widely studied and PF algorithm by simulation in the univariate non-static growth model.The advantages and disadvantages of the algorithms are analyzed and compared with the classic filtering algorithms.The results show that the algorithm presented in this thesis has higher accuracy. |
