| 摘要: |
| 为解决虚假目标点迹对雷达跟踪性能的影响，本文提出了一种基于PSO-SVM算法的雷达点迹真伪鉴别方法，进一步对目标点迹和杂波点迹进行真伪鉴别，有助于滤除杂波剩余点迹，提高雷达处理容量和跟踪性能。本方法利用点迹形成过程中生成的特征参数，先利用PSO算法对SVM算法参数进行优化选择，再利用参数优化后的SVM算法对雷达点迹进行真伪鉴别。最终，目标点迹鉴别准确率达到了95.18%，杂波点迹鉴别准确率达到了89.94%，整体的点迹鉴别准确率达到了92.13%。实验结果表明：该算法有较高、较稳定的点迹鉴别准确率，前期较多的杂波点迹被鉴别为目标点迹的缺陷也得到了较好的改善。 |
| 关键词: 剩余杂波 支持向量机 粒子群算法 点迹鉴别 |
| DOI：DOI:10.3969/j.issn.1672-2337.2020.04.012 |
| 分类号:TN957 |
| 基金项目: |
|
| An Identification Method of True and False Plots Based on PSO-SVM Algorithm |
|
PENG Wei， LIN Qiang
|
|
Air Force Early Warning Academy， Wuhan 430019， China
|
| Abstract: |
| In order to remove the influence of false target plot on radar tracking performance, this paper proposes an identification method of true and false plots based on PSO-SVM algorithm. With this method, further identifications of the true and false plots and clutter plots are helpful to filter residual clutter plot and improve radar processing capacity and tracking performance. This method uses the characteristic parameters gene-rated in the process of plot formation. First, the PSO algorithm is used to optimize the parameters of SVM algorithm. Then, the optimized SVM algorithm is used to identify the true or false plots. Finally, the accuracy rate of target plot and clutter plot identification can reach up to 95.18% and 89.94% respectively.The overall accuracy rate of plot identification can reach 92.13%. The experimental results show that the algorithm has a higher and more stable accuracy rate of plot identification and the early defect of clutter plots being identified as target plots has also been improved. |
| Key words: residual clutter support vector machine(SVM) particle swarm optimization(PSO) plot identification |
