Information Fusion Technology For Underwater Target Recognition

Multisensor data fusion is a rising discipline, and has wide applications in areas of radar, sonar, navigation, automatic control, communication and robot. Data fusion techniques combine data from multiple sensors or multiple kinds of sensors, and related information from associated databases, to achieve a better performance of detecting, tacking, classifying targets than performance could be achieved by the use of a single sensor alone .The distributed data fusion can rise the system's reliability and survivability effectively.In the dissertation, the basic theories and fusion algorithms of data fusion and its applications in object identification system is studied .The main contributions of this dissertation are summarized as follows:Systemically introduces the basic concept of technology -IF (information fusion). Additionally, the classification of system, the kinds of system modeling, algorithm for identification and introduce the rule of IFC(infbrmation fusion center) are discussion in detail.Particularly discusses Bayesian fusion and Dempster-Shafer Method used to multisensor signal processing. Simulate at the target of the underwater for the naval vessel platform. The result of simulation reveals that it has better performance by the use of multisensor than by the use of a single sensor alone. Additionally, compares the different Bayesian fusion and Dempster-Shafer Method. The comparison illustrates the advantage and advance.Introduces in detail data fusion for multisensor target state estimation in view of Kalman filtering. Concentrate on average fusion and hierarchical fusion in the area of the detonator.Bring forward the fuzzy model and algorithms for data fusion. In simulation, the admission function is first defined to measure the reliability of sensors. The final result is therefore the combination of compatible data provided by more reliable sensors.