| 摘要: |
| 针对雷达信号分选中辐射源数目不确定、脉冲数据分布形式复杂、对噪声影响敏感等问题，提出了一种基于改进谱聚类联合数据场理论的聚类分选算法。该算法首先利用数据场理论对数据进行预处理，根据势值大小实现干扰点的去除，并利用势心的数目确定初始聚类数，然后再利用网格密度划分得到合理的地标点，最后再基于地标稀疏表示的谱聚类算法完成聚类分选。通过设置两组类型不同的脉冲信号数据进行仿真实验，分选正确率均达到95%以上，验证了该算法具有较高准确率和鲁棒性。 |
| 关键词: 雷达信号分选 谱聚类 数据场 网格密度 稀疏表示 |
| DOI：10.20079/j.issn.1001-893x. 220425001 |
|
| 基金项目: |
|
| A Radar Signal Sorting Algorithm Based on Improved Spectral Clustering |
| WANG Yili,YANG Yuming |
| (School of Mathematical Sciences,University of Electronic Science and Technology of China,Chengdu 611731,China) |
| Abstract: |
| For the problems such as uncertain number of emitters,complex distribution of pulse data and sensitivity to noise influence in radar signal sorting,a cluster sorting algorithm based on improved spectral clustering combined with data field theory is proposed. The algorithm first uses the data field theory to preprocess the data,removes interference points according to the size of the potential value,and uses the number of potential centers to determine the initial number of clusters.Then it uses the grid density to get reasonable landmark points,and finally completes the cluster sorting based on the spectral clustering algorithm of landmark-based sparse representation. Two sets of different types of pulse signal data are set up for simulation experiments and the sorting accuracy reaches more than 95%,which verifies that the algorithm has high accuracy and robustness. |
| Key words: radar signal sorting spectral clustering data field grid density sparse representation |
