Study Of Radar Target Tracking Methods After Video Level Fusion

In this paper, an introduction about the basic principles of data association and target tracking is given. The theory and the process of data association are explained. Several common data association algorithms are researched deeply. Their principles and realizations are analyzed. The advantages, disadvantages and application are demonstrated. Then a simulation of nearest neighbor data association is made in this paper.For the multi-radar data which is processed by video level fusion, a method of track association based on the estimate of the speed and the course is proposed. The new method depends on the data which was processed by video level fusion in multi-radar system and consists of two parts which are the estimate of the speed and the estimate of the course. By using the range of the speed and the course after gate association, the wrong points are picked out and the right point is found. It ensures that the point which is associated is in line with the common target. The simulation shows that this method is in good performance and easy to apply in engineering. Finally the adaptive Kalman filtering algorithm and the adaptiveα-βfiltering algorithm are compared by Monte Carlo simulation. From the result the differences of precision and efficiency between the adaptive Kalman filtering algorithm and the adaptiveα-βfiltering algorithm are analyzed.