Research On Tracking Filtering Algorithm And Its Application In Target Tracking Using High Frequency Surface Wave Radar

High Frequency Surface Wave Radar is an important detection tool of ship targets on sea surface witch has the ability to monitor targets in a large area and any weather. Ship target navigation tracking is quite a significant process of target detection. It enables High Frequency Surface Wave Radar to continuously detect, eliminate false target, and improve detection accuracy as its filtering function.High Frequency Surface Wave Radar target navigation tracking involves many procedures such as track initiation, data association, tracking filter, etc. Among them, tracking filter is a very important process. At present, tracking filtering methods using in High Frequency Surface Wave Radar target tracking mainly include Kalman filtering and alpha beta filter. As real data is achieved under the polar coordinate, tracking in polar coordinate will evade coordinate transformation. In this paper, alpha beta filter based and Kalman filter based High Frequency Surface Wave Radar target navigation tracking methods are studied respectively under polar coordinates. Tracking performance is evaluated combined with real measured data. Main work of this paper is as follows:(1) A simulation model is proposed for carrying out navigation tracking simulation experiments under polar coordinates. As AIS data has different time and spatial scale, interpolation process is imposed to get synchronizing data matching in time and space for result evaluation. Corresponding method is given for each link of the whole process.(2) An adaptive alpha beta filtering algorithm is proposed as the common alpha beta filtering algorithm has a const filtering coefficient which is not suitable in any radar data process circumstances. This method establishes a function to get an adaptive filtering coefficient using velocity, range and angle deviation. In addition, filtering coefficient priority is allocated as the fact that HFSWR has a higher velocity measurement and a worse angle measurement. Simulation experiment results and AIS information based measured data processing results both show that the proposed method has a low calculation time and higher accuracy, in addition the proposed method is able to track remote target navigation.(3) Noise variance is necessary in achieving Kalman filtering gain in polar coordinate, yet it is difficult to determine in real circumstance. For this purpose, a new Kalman filtering gain acquisition method is proposed in this paper. This method enables a more accurate noise variance at a certain sample time using filtering result and measured data. The filtering gain will be more accurate and the tracking result will be better based on the noise variance. Simulation experiment result and AIS information based measured data processing result both show that the proposed method will get higher tracking accuracy and a better tracking result of remote navigation. Further more, comparative analysis of alpha beta filtering and Kalman filtering is carried out. Result shows that the former using less time resource and the later has higher accuracy. In real application, a selection from the two methods will be made based on certain processing demand such as a higher precision or a faster process.