Research On Direct Position Determination And Tracking Technology Of Radiation Source Targets

The positioning and tracking technology for radiation source targets is an important component of electronic reconnaissance systems.The traditional radiation source target localization and tracking technology adopts a two-step method,which first estimates the positioning parameters such as time difference,frequency difference,and direction using signals intercepted by multiple stations.Then,based on this,further localization and tracking calculations of the radiation source target are carried out.Traditional methods have poor performance under low signal-to-noise ratio conditions due to the accumulation of parameter errors.The direct positioning and tracking technology of radiation source targets directly utilizes the intercepted signal data itself to obtain spatial information of radiation source targets through maximum likelihood and Bayesian theory analysis.Compared to traditional two-step methods,it has better performance under low signal-to-noise ratio conditions and has received widespread attention in related fields.This thesis focuses on the direct positioning and tracking technology of radiation source targets,and the main content is as follows:1.Introduced the traditional two-step localization method based on arrival angle and time-frequency difference information.We introduced random finite sets and Bayesian filtering,as well as three common filters based on Bayesian principles and their predictive update formulas.2.In response to the performance degradation of traditional two-step localization methods under low signal-to-noise ratio,a direct localization technology based on passive systems was studied.Firstly,the direct localization technology of radiation sources based on time-frequency difference was studied,and the maximum likelihood principle was used to achieve direct estimation of the radiation source state.The Cram é r Rao Bound(CRB)in this scenario was derived.Based on this,a direct localization algorithm based on spread spectrum code sequence estimation was studied for direct spread spectrum sequence radiation source targets.The algorithm achieved direct localization of the radiation source on the basis of blind signal despreading.Its performance in low signal-to-noise ratio environments was verified through simulation experiments.3.A particle filter direct tracking algorithm based on time-frequency difference was studied using Bayesian filtering principle to address the problem of state estimation for moving single radiation sources.By constructing a fuzzy function as likelihood and modeling it with probability,the state estimation of moving radiation sources is achieved.To solve the problem of detection errors under low signal-to-noise ratio,a joint detection and direct tracking algorithm based on multiple arrays was studied by introducing the idea of random finite sets.Probability modeling of the sampling covariance of the signal,utilizing the arrival angle information of the radiation source,and using Bernoulli filters to achieve detection and state estimation of the target radiation source.Simulation experiments have shown that it has excellent detection and estimation performance in low signal-to-noise ratio environments.4.A multi array based joint detection and direct tracking technology for multiple radiation sources is studied based on the superposition measurement model generated by multiple radiation sources to address the problem of state estimation for moving multiple radiation sources.Using the arrival angle information of multiple radiation sources,using Wishart and inverse Wishart distributions,probability modeling is performed for sampling covariance.Then,a potential probability hypothesis density filter is used,combined with an auxiliary particle filter algorithm,to estimate the state of multiple radiation sources.Simulation experiments have shown that it has excellent detection and estimation performance in low signal-to-noise ratio environments.