基于多维加权聚类的雷达信号分选方法

随着数字技术的发展,新体制、新技术雷达不断出现,反辐射导引头面临的信号环境更加密集和复杂,传统的方法已经不能对其进行有效分选。文中提出一种新的基于多维加权聚类的雷达信号分选算法。通过统计和熵值分析确定权值,计算加权的欧几里得距离,根据既定门限合并同类聚类中心,从而实现雷达脉冲的分选,并在各聚类中心应用改进的SDIF算法进行PRI精分选。计算机仿真结果验证了该方法分选的有效性。

Radar Signal Sorting Method Based on Multi-dimension Weighed Clustering

The new system and new technical radar is occurred continuously with the development of digital techniques,so the signal environment that Anti-radiation Missile(ARM) faces is more and more intensive and complicated.The traditional methods can't sort radar signals effectively.In this paper,a new radar signal sorting algorithm based on multi-dimension adaptive weighed clustering is introduced.The weighed values are determined based on the statistics and entropy theory,the weighted Euclidean distance is calculated,and the similar clustering centers are merged according to the fixed threshold.Thus radar signals are sorted into individual clusters,and further sorting is done by using SDIF algorithm according to pulse repeat interval(PRI) in every clustering.The computer simulation results show that the proposed method can effectively sort radar signals.