Non-Cooperative Target Recognition The Technology Of Data Fusion

The paper elaborates the principle on D-S multi-sensors data fusion and target recognition. At last, an example is given. How to use Dempster-Shafer (D-S) method to solve multi-sensor data fusion problems is analyzed in this paper. Based on basic probability assignment of target type decided by multiple sensors, new sensor data are added continually, and believe function and plausibility function are update; finally the destination of decision of target type is arrived.Several results of Dempster-Shafer (D-S) method applied to target fusion identification of two sensors and conclusions drawn from them were given and proved. The recurrence formula of multi-sensor (>2) D-S fusion identification were derived and their properties were shown clearly. The study provides theoretical basis for selecting various sensors and effectively fusing information from them in multi-sensor target identification systems. Furthemore, the recurrence formula of fusion identification can reduce computational load, enhance adjustability of distributed multi-sensor identification systems, and they can be used in the case of real time. These results have great value for solve problem of military target identification.