Based Particle Filter Single Observer Passive Location Tracking Technology

Target localization and tracking is applied in the system of guidance, surveillance, and obstacle avoidance, whoes role is to determine position, movement, and identity of targets. Passive localization and tracking system plays an important role in the electronic reconnaissance, because it works silently without any electromagnetic radiation. And since single observer passive localization and tracking (SOPLAT) technology avoid time synchronization and communication among observers, it becoming more and more popular in target localization field.Essentially, It is more important to choose a right filtering algorithm in single observer passive localization and tracking,because filtering technology is one of important technology and it can be direct to effect SOPLAT. Recently there has been a surge of interest in nonlinear and non-Gaussian filtering algorithm, particle filtering(PF) algorithm is much payed attention to naturally. According to characteristic of SOPLAT, it is very important to research PF algorithm used in the SOPLAT. Thus, the dissertation does deeper researches on the PF algorithm used in the SOPLAT.First of all, the passive localization technology is briefly analyzed, including typical localization methods and filtering algorithms. In Chapter 2, the mathematical model of localization system is set up. Based on this model, the principle of many kinds of localization methods is discussed.In Chapter 3, the Bayesian inference is discussed at first. Then the filtering algorithms of SOPLAT are mainly discussed. EKF and UKF algorithm is analyzed, EKF and UKF algorithm is the most classical nonlinear method in SOPLAT, successfully applying in many passive localization problems. The computer simulations are carried out, and the performance and restrict in practical application is analyzed.In order to avoid the weakness of classical method. PF algorithm is brought out. In Chapter 4, fundamental principle of PF algorithm is studied at first. Then, SIS framework is given, importance density choice is mainly studied. At last the generic particle filtering(GPF) algorithm framework is given and resampling is simplely discussed.And then, Sequential Importance Resampling Particle Filtering(SIRPF) algorithm, Extended Kalman Particle Filtering(EKPF) algorithm, Unscented Particle Filtering(UPF) algorithm and their application in SOPLAT are mainly studied. The computer simulations of these algorithms with different locating method are carried out. At last, we use EKF, UKF and PF algorithm to tracking a single target. The result show the PF algorithm has its superiority over the other two methods. And we compare them from linearity and distribution.Finally, chapter five summarizes the main benefits,challenges and its future of partilce filtering algorithm.