摘要
针对新型作战样式条件下空中多机动目标密集回波的数据关联问题,采用核学习方法和K-均值聚类相结合的算法,即基于核的K-均值聚类来解决此问题。该方法的主要思想是,将原空间中的样本通过一个非线性映射,映射到高维的核空间中,以突出各类样本之间的特征差异,然后在核空间中进行K-均值聚类。仿真结果表明,该方法有效提高了密集回波环境下系统跟踪机动多目标的关联精度和可靠性。
Aiming at the data association problem in high dense multi-return environment, kemel learning method and K-means algorithm, namely the K-means cluster based on kernel, are combined to solve the data association of multi-maneuvering targets. The idea of this algorithm is mapping the sample from the original space to a high dimension kemel space where there are characteristic differences among allkindsofsamples, and then perform K-means clustering in this high dimension space. The simulation results indicate that this method can effectively improve the precision and reliability of the system under dense multi-return environment.
出处
《计算机工程与设计》
CSCD
北大核心
2007年第20期4845-4846,4849,共3页
Computer Engineering and Design
基金
国防预研应用基础研究基金项目(A1420061266)
关键词
K-均值聚类
核聚类
核函数
机动多目标
数据关联
k-means cluster
kernel cluster
kernel function
multi-maneuvering targets
data association
