一种新的k-means聚类雷达信号分选算法

发布时间：2018-03-28 07:09

本文选题：雷达信号分选　切入点：k-means聚类　出处：《现代防御技术》2015年06期

【摘要】：针对传统k-means聚类算法在雷达信号分选中应用存在的不足,提出了一种基于数据场和灰关联分析的k-means聚类雷达信号分选算法。该算法首先根据数据场理论计算所有数据样本的势值,寻找局域势值最大值,选取距最大值最近的样本数据作为初始聚类中心,局域势值最大值个数作为聚类数目;然后用灰关联度代替欧式距离来判断数据样本间相似性。该算法能够自动获取初始聚类中心和聚类数目,对频率捷变雷达具有较好的分选效果。仿真结果验证了算法的可行性。
[Abstract]:Aiming at the shortcomings of traditional k-means clustering algorithm in radar signal sorting, A k-means clustering radar signal sorting algorithm based on data field and grey correlation analysis is proposed. Firstly, the potential value of all data samples is calculated according to the theory of data field, and the maximum local potential value is found. The sample data closest to the maximum value is selected as the initial clustering center, and the maximum number of local potential values is taken as the clustering number. The grey correlation degree is used instead of the Euclidean distance to judge the similarity of data samples. The algorithm can automatically obtain the initial clustering center and the number of clusters, and has a good sorting effect on frequency agile radar. The simulation results show that the algorithm is feasible.
【作者单位】：南京信息工程大学电子与信息工程学院;中国船舶重工集团第七二四研究所;
【分类号】：TN957.51

【参考文献】