辐射源识别已成为军事情报、监视、侦察系统中的关键问题。在战场环境中,多传感器探测到的辐射源信息呈现出不确定性、不完整性和矛盾性。本文提出了一种基于粗糙集和D-S推理的多传感器辐射源识别模型,并提出了一种新的基于分类质量的属性约简算法,仿真分析验证了该模型的有效性。 Radiation source identification has become a key issue in military intelligence, surveillance and reconnaissance systems. In the battlefield environment, the information of the radiation source detected by the multi-sensor shows the uncertainty, the incompleteness and the contradiction. In this paper, a multi-sensor radiation source identification model based on rough sets and D-S inference is proposed, and a new attribute reduction algorithm based on classification quality is proposed. The simulation results verify the effectiveness of the model.