Research On Data Fusion Technique Of Multi-sensor System

In modern C3I(Command,Control,Communication and Intelligence)systems, information provided by single sensor can't be satisfied by the need of modern war. Multi-sensor must be used to measure data, detect and identify object in real time. Therefore technology of multi-sensor data fusion developed rapidly, and was widely applied to modern C3I systems and weapons. Presently, the data fusion technique has become a very active research field.Network Weapon System(NWS) composes a group of micro-attacking units which can identify, locate, and attack the targets. It is a intelligent weapon system. In order to improve the accuracy of target identification, the multi-sensor data fusion technology is used in NWS. Multi-sensor data fusion includes a number of methods. Compared with other data fusion methods, evidence reasoning theory makes decisions according to human's way, and can be explained reasonably by using information theory. Certain and uncertain zones can be divided with no prior information and randomness, and uncertainty from imprecise sources can be managed intuitively. Evidence reasoning is widely used in dealing with the uncertain information.The thesis systematically presented data fusion technology, such as basic principles, the sense of study, the situation of study in domestics and abroad, the current study emphases and it's development. In addition, we also discussed the system model, structure and main algorithms of data fusion technology. In the thesis a method of solving the problem of uncertainty——Dempster-Shafer evidence reasoning was studied principally,and was applied to NWS.Finally, the data is calculated based on the mathematics models. The result proved that the mathematic models can improve the precision of target recognition, and the data fusion technology can be used in NWS successfully.