Study Of UAV Passive Localization And Tracking Using Multipath Information

Recently with the spreading usage of UAV,the violate flights and flight accidents increase accordingly.There is an increasing demand on localizing and tracking the UAV.Whereas,UAV has its unique feature as its small shape and relatively low velocity and the radio control signal is on certain frequency,w hich makes it is hard to be detected b y radar.Therefore,TDOA based localization and tracking methods can be applied.In this paper,traditional TDOA based methods are introduced,it can be divided into two steps: TDOA estimation and localization solving.Under multipath environment,the accuracy of TDOA estimation can not be guaranteed.Furthermore,traditional methods can not deal with the situation of dynamic targets number.Thus,in this paper,Random Finite Set based target tracking method is used.For UAV localization and tracking,it can be viewed as a Markov process and the moving velocity is low.Thus,target tracking algorithm in Bayes framework can be used to the localization problem.The typical Probability Hypothesis Density filter is used and extended to a multipath and multisensor version.Take the advantage of the multipath information during the localizing and tracking UAV process.Also,under the urban area multipath environment,the multipath information can increase the observability of the targets.In this paper,the multipath information is used by an extended version of PHD filter,i.e.multi-detection PHD filter.Meanwhile,taking the non-linear TDOA equations,particle filter can be applied to implement the multi-detection PHD algorithm.The perdition process is the same with the traditional PHD filter,whereas the update process is different.Due to the uncertainty of the relationship between the observations and the observation models,the pseudo likelihood function will consider all cases.By such iteration,the target state can be obtained by clustering method from the updated PHD.Finally,three test scenarios are promoted.They are single station with multipath,multi-station with single path and multi-station with multipath.Algorithms are tested under the self-developed software platform for target localization and tracking.Simulations show that using multipath information can improve the accuracy of UAV target localization and tracking.