Abstract: Although the different type stations have the same tactical parameters，the technical parameters of them are different because of communication mode.Therefore，this paper identifies the stations by considering the parameters separately and makes the identification by matching the technical parameters，which is based on matching the tactical parameters.Rough set theory is applied to reduce the attributes and extract the main attributes，then the weight is automatically set according to the attributes by calculating the importance of them.The results of the simulation experiments have demonstrated the efficiency of this algorithm.

Key words: radio station identification, rough set, weight

摘要： 不同类型的电台可以有相同的战术参数，但技术参数由于通信方式的不同会有较大的差异，因此分战术参数和技术参数分层识别电台。在战术参数识别的基础上，进行技术参数匹配，实现对电台整体识别。两层识别中，均利用粗糙集理论进行属性约简，提取核心属性，计算其重要性，以此自动设置各核心属性的权系数。仿真结果表明该算法是有效的，能提高电台识别的正确率。

关键词: 通信电台识别, 粗糙集, 权重