摘要
空间运动目标RCS数据序列能反映出空间目标的姿态运动特征。针对空间运动目标RCS数据序列的变化规律,首先仿真生成低速旋转空间目标的RCS数据序列,而后采用小波变换、傅里叶变换以及RCS数据序列统计学特征提取等方法,对低速旋转空间目标的RCS数据序列进行特征提取。最后采用朴素贝叶斯、支持向量机、随机森林分类和logistic逻辑回归算法等机器学习分类算法,实现了对低速旋转空间目标RCS数据序列的识别。
The RCS sequence of space moving target can reflect the attitude characteristics of space target.According to the change law of RCS sequence of space moving target,this paper first simulates and generates the RCS sequence of low-speed rotating space target,and then uses the methods of wavelet transform,Fourier transform and statistical feature extraction of RCS sequence to extract the features of RCS sequence of low-speed rotating space target.Finally,machine learning classification algorithms such as naive Bayes,support vector machine,random forest classification and logistic regression algorithm are used to identify the RCS sequence of low-speed rotating space targets.
作者
于兴伟
张学文
侯鑫宇
张超
YU Xingwei;ZHANG Xuewen;HOU Xinyu;ZHANG Chao(The Unit 95921 of PLA,Jinan 250000,China)
出处
《现代雷达》
CSCD
北大核心
2022年第7期75-81,共7页
Modern Radar
关键词
RCS数据序列
低速旋转空间目标
机器学习分类算法
RCS sequences
low speed rotating space target
machine learning classification algorithm
