摘要
为有效解决脉冲信号分选中辐射源数量不确定和脉冲数据分布形式复杂的问题,文中提出了一种基于仿射传播聚类的辐射源脉冲信号分选算法。该方法通过构造脉冲信号样本数据的拉普拉斯矩阵,并计算其矩阵的特征向量作为参数空间变换后样本新特征向量,然后采用仿射传播聚类算法对新样本向量进行聚类,最终实现辐射源脉冲信号分选。经过仿真试验验证了该算法对于未知辐射源脉冲信号分选的有效性,辐射源脉冲信号分选正确率达到94.47%,试验结果表明了文中所提算法克服了辐射源数量不确定和信号样本复杂分布的问题,对于辐射源脉冲信号分选具有较高准确性和鲁棒性。
To solve the problems of the unknown number of the emitters and the complex distribution of data,a new emitter signal sorting method based on the Affinity Propagation clustering is proposed in this paper.It makes the Laplace matrix of the pulse signal data and calculate the eigenvector of the matrix as the new characteristic of the signal data,and then it applies the affinity propagation clustering to signal sorting.The simulation results show that the proposed approach is effective to the unknown emitter signal sorting and the correct sorting rate of emitter signal data is 94.47%.The new method proposed in this paper overcomes the problems of the unknown number of emitters and the complex distribution of data,which also has good results of the unknown emitter signal sorting with high accuracy and robustness.
作者
何重航
刘佳琪
王宪栋
刘洪艳
高路
HE Zhong-hang;LIU Jia-qi;WANG Xian-dong;LIU Hong-yan;GAO Lu(National Key Laboratory of Science and Technology on Test Physics and Numerical Mathematic,Beijing 100076 China;The Unit 61336 of PLA,Beijing 100094 China)
出处
《中国电子科学研究院学报》
北大核心
2020年第10期1024-1030,共7页
Journal of China Academy of Electronics and Information Technology
关键词
信号分选
仿射传播聚类
参数空间变换
谱聚类
signal sorting
affinity propagation clustering
parameter space transformation
spectrum clustering
