Research Of Multisensor Detection & Position Level Fusion In Intelligence Analysis Module In Electronic Warfare System

The intelligence analysis systems is an important part of the Electronic Warfare(EW) systems, needed processing different information comes from different sensors. Data comes from Radar, Electronic Support Measures(ESM), Infrared Sensor, Photovoltaic Radar, Spy Satellite will contain different information or the same information. So, it is very important for Electronic warfare systems to find which information is important and which is useless. Systems will get a right Situation awareness and threat assessment. In order to accomplish this task, the accuracy and integrity of information is important. Therefore, the modern EW system need to face the ?big data? problem. It?s important to find an effective way to fusion those different message comes from different platforms. At the same time, some of those messages are incomplete. Therefore, this thesis will focus on the different source image fusion technology, different source track association technology and track fusion technology. In order to show the works of this paper, four points in this paper will be proposed in next four paragraphs.(1) This thesis discussed the important of the multi sensor data fusion technology in modern EW system. At the same time, this paper discussed the development history of the image fusion technology and the track fusion technology.(2) This thesis proposed a SAR-Optical image fusion method based on dual-tree complex wavelet transform(DT-CWT), Simulation result by SAR-Optical image fusion in urban area shows the effective of this algorithm.(3) This thesis proposed a Rader-ESM track correlation algorithm based on the optimal Bayes filter and discussed the shortcoming of the classical Radar-ESM track correlation algorithm. An ?association-while-fuse? structure have been proposed in this paper. Simulation result shows the effective of this algorithm.(4) By analysis the Kalman filter(KF), this thesis proposed a track fusion method based on the probability of the measurement. Which calculate the weight by calculate the probability of the measurement in k-th time. Also, an optimal fusion method by calculate the virtual measurement in out-of-sequence track fusion have been proposed. Simulation result shows the effective of those two algorithms.