一种基于加权证据合成的多传感器目标识别方法

针对多传感器数据融合目标识别问题，基于D-S证据理论，提出了加权证据合成的时空域目标识别算法。该方法充分利用了多传感器多周期的测量数据，并根据D-S合成规则要求参与合成的各证据具有相同权重的特点，充分考虑了提供证据的信源即各个传感器的可靠性。在合成中，引入证据权的概念，解决了不同权重的多传感器数据融合问题，在一定程度上改善了目标识别系统的性能。最后通过计算实例表明算法是有效的。

Multisensor target identification method based onweighted evidence combination

To tackle target identification problem of multisenor data fusion, a temporal-spatial target identification algorithm with weighted evidence combination was proposed based on D-S evidence theory. The method adequately uses the multicycle measurement of multisensor. And according to the character that D- S rule requires all evidences being combined should have the same weight, the method also considers the reliability of every sensor. By introducing the concept of evidence weight, the method solves the data fusion problem of multisenor with different weight and then improves the performance of target identification system to a certain extent. Finally an example illustrates the validity of the algorithm.