A Research Of Multi-target Tracking Method With Asynchronous Multi-station Radar Systems

In the increasingly complex modern battlefield environment,the multi-station radar system has been widely concerned and applied due to the significant advantages in expanding the spatial and temporal coverage,reducing the ambiguity of data,and improving the robustness of the system.Especially,target tracking is a critical element in the war.Thus,it has been a hot research topic to use the multi-station radar system tracking targets.However,due to the different initial sampling time and sampling rates of the local radars and the random time delay during the information transmission,the multistation radar system using for target tracking are asynchronous.Aiming at the target tracking in the asynchronous multi-station radar system,this thesis focuses on the asynchronous data fusion of in-sequence measurements and the asynchronous data fusion of out-of-sequence measurements.The main contributions are as follows:1.The network topology and the system model of the multi-station radar system are analysed.Then,the target tracking algorithms based on Bayesian filter are studied,which provides a theoretical foundation for the target tracking methods using the asynchronous multi-station radar system in the subsequent chapters.2.Aiming at the multi-target tracking using the in-sequence measurements in the asynchronous multi-station radar system,an asynchronous periodical sequential fusion algorithm based on probability hypothesis density(PHD)filter is proposed.By jointly using asynchronous in-sequence measurements in an update periodical to sequentially update the target state in the update time,the target tracking in the asynchronous multistation radar system is achieved.Besides,jointly using the measurements in an update periodical can improve the tracking accuracy.3.Aiming at the single-target tracking using the out-of-sequence measurements in the asynchronous multi-station radar system,a single-target tracking algorithm using the out-of-sequence measurements based on the particle filter(PF)is proposed.By deriving the update expressions using the out-of-sequence measurements in the framework of Bayesian theory,the single-target tracking method using the out-of-sequence measurements in the arbitrary orders is provided,which solves the update problem of updating the out-of-sequence measurements in the arbitrary orders.4.Aiming at the multi-target tracking using the out-of-sequence measurements in the asynchronous multi-station radar system,a multi-target tracking algorithm based on probability hypothesis density filter is proposed.The proposed algorithm consists of two stages: retrodiction and posterior update,which solves the problem of the mismatched targets information contained by the out-of-sequence measurements and the filtering posterior intensity function.Also,through the two stages of the proposed algorithm,the multi-target tracking using out-of-sequence measurements in the asynchronous multistation radar system is achieved,and the target tracking performance is improved.The above works have been demonstrated by the simulation,and the numerical results show the effectiveness of the proposed algorithms.