Research On Target Tracking And Recognition Technology In Data Fusion

With the development of science and technology, sensors greatly improve, therefore, multi-sensor information systems for various complex application background are came forth and widely applied in the military field as well as many civilian areas. Multi-sensors data fusion technology synthetizes datum of many sensors, thus draws more accurate and reliable conclusions, moreover, comparing with the data of any simple sensor, it supplies more exactitude and accuracy consequences. The scope of this paper is the military data fusion which is an important part of C3I system. The function model of data fusion includes signal detection, position estimation, identify estimation in the low level and situation estimation, threat estimation in the high level.This paper focuses target tracking and targets recognition in the military data fusion. The work in these two areas as follows.Firstly, when multiple targets are tracked, by the theoretical foundation of multi-sensor data fusion, multi sensor joint probability data internet algorithms which are based on single sensor data joint probability is introduced. Sensor data processing is put forward and goals movement model is built, multi sensor data internet algorithms and Kalman filter are focused on. Under MATLAB environment, two trajectories meet goals are tracked and satisfactory simulation results are achieved.Secondly, after the accurate tracking of targets, passive detection sensors are used for identifying emitters of the target platform. The relations table of radiation sources and platforms are established and stored. Furthermore, Fuzzy production rules are utilized to reasoning the target platform. Finally, D-S evidence theory is used for fusing the information of sensors, Therefore, The platform is identified.