电光与控制, 2020, 27 (4): 82, 网络出版: 2020-12-08  

雷达干扰信号识别决策树的自动化设计方法

A Method for Automatic Design of Decision Tree in Radar Jamming Signal Recognition
作者单位
南京航空航天大学, 南京 211100
摘要
针对雷达干扰识别决策树分类器设计需要人工介入的问题, 提出了一种基于模糊聚类、Xie-Beni指标和信息增益的决策树自动化设计方法。该方法首先对干扰信号在时域、频域和脉压后时域等维度建立参数特征集, 接着在决策树建立过程中引入模糊C均值聚类(FCM), 从而解决传统决策树需要先验知识设置判决门限的问题; 然后通过Xie-Beni指标动态确定决策树节点分支数, 优化决策树复杂程度; 最后使用基于信息增益的ID3算法建立模糊聚类决策树。该方法解决了干扰识别决策树的自动化设计问题, 且优化了决策树性能。计算机仿真及某雷达对抗实验数据验证了所提方法的有效性。
Abstract
To solve the problem that manual intervention is required in the design of the classifier of the decision tree in radar jamming signal recognition, a method for automatic design of the decision tree based on fuzzy clustering, Xie-Beni index and information gain is proposed.Firstly, the feature set of the parameters of the jamming signal in such dimensions as the time domain, the frequency domain and the time domain after pulse compression is established.Then, Fuzzy C-Means (FCM) clustering is introduced in the process of decision tree establishment, so as to solve the problem that the traditional decision tree needs prior knowledge to set the decision threshold.Then, Xie-Beni index is used to dynamically determine the number of branches of decision tree nodes and lower the complexity of the decision tree.Finally, the ID3 algorithm based on information gain is used to establish the fuzzy-clustering decision tree.This method realizes the automatic design of the decision tree in jamming signal recognition and optimizes the performance of the decision tree.Computer simulation and the experimental data of radar countermeasure have verified the effectiveness of this method.

魏煜宁, 张劲东, 李勇, 苟立婷. 雷达干扰信号识别决策树的自动化设计方法[J]. 电光与控制, 2020, 27(4): 82. WEI Yuning, ZHANG Jindong, LI Yong, GOU Liting. A Method for Automatic Design of Decision Tree in Radar Jamming Signal Recognition[J]. Electronics Optics & Control, 2020, 27(4): 82.

关于本站 Cookie 的使用提示

中国光学期刊网使用基于 cookie 的技术来更好地为您提供各项服务,点击此处了解我们的隐私策略。 如您需继续使用本网站,请您授权我们使用本地 cookie 来保存部分信息。
全站搜索
您最值得信赖的光电行业旗舰网络服务平台!