摘要
针对传统方位聚类算法由于采用固定门限而难以准确对密集方位目标进行准确分选的难题,提出一种基于重频熵的自适应方位聚类算法。算法首先定义了重频熵数学模型,然后采用一系列不同的方位门限对脉冲进行聚类,并依据重频熵对不同方位门限的脉冲聚类结果进行度量,最后选取最优方位门限对脉冲进行聚类处理。仿真和实际数据验证表明,本算法可处理得到准确的聚类结果,与传统聚类算法相比本算法在密集信号环境下的聚类准确率可提升15%以上。
Aiming at the difficulty of accurately sorting dense azimuth targets in traditional azimuth clustering algorithms due to the use of fixed thresholds,a self-adaptive azimuth clustering algorithm based on pulse repetition interval(PRI)entropy is proposed.Firstly,the mathematical model of PRI entropy is defined by the algorithm.Then,a series of different azimuth thresholds are used to cluster the pulses.The pulse clustering results of different azimuth thresholds are measured based on PRI entropy.Finally,the optimal azimuth threshold is selected to cluster the pulses.Simulation and actual data validation show that this algorithm can process accurate clustering results,and the clustering accuracy of this algorithm in dense signal environment can be improved by more than 15%compared with traditional clustering algorithms.
作者
沈路
褚心童
杜冶
杨启伦
赵巍
SHEN Lu;CHU Xintong;DU Ye;YANG Qiun;ZHAO Wei(Southwest China Research Institute of Electronic Equipment,Chengdu 610036,China)
出处
《电子信息对抗技术》
2024年第2期40-45,共6页
Electronic Information Warfare Technology
关键词
重频熵
方位聚类
信号分选
自适应
脉冲聚类
PRI entropy
azimuth clustering
signal sorting
self-adaptive
pulse clustering
