| 本文已被：浏览 435次 下载 0次 |
|
|
|
| 基于CNN-BiLSTM的对ELINT系统杂乱脉冲干扰 |
|
鲁永为1,2, 师俊朋2, 周青松2, 田西兰3, 郭柏炀2, 陈沁娴2, 山世浩1, 王明壮4
|
|
1.解放军63891部队 洛阳;2.国防科技大学电子对抗学院 合肥;3.中国电子科技集团公司第三十八研究所 合肥;4.解放军96816部队 启东
|
|
| 摘要: |
| 针对杂乱脉冲干扰信号严重破坏电子情报系统对雷达信号的分选识别问题，将机器学习算法引入侦察分选过程，提出一种基于卷积神经网络—双向长短期记忆网络的抑制杂乱脉冲干扰方法。该方法利用卷积神经网络捕捉杂乱脉冲干扰信号在多个维度随机分布的局部特征，以及双向长短期记忆网络网络捕捉混叠信号中雷达信号的周期特征，实现干扰信号和雷达信号的分类识别。仿真实验证明该方法针对不同的测试集准确率达到96%以上，召回率超过92%，有较强的域泛化能力，为提升ELINT系统在复杂电磁环境下的性能提供了一种有效途径。 |
| 关键词: 电子情报系统 卷积神经网络 双向长短期记忆网络 杂乱脉冲干扰 雷达侦察 |
| DOI： |
| 分类号:TN95 |
| 基金项目:国家自然科学基金(No. 62301581，62401600)，中国博士后科学基金面上项目(No.2023M734313) |
|
| Method for Suppressing Chaotic Pulse Interference in ELINT System Based on CNN-BiLSTM |
|
|
|
|
| Abstract: |
| In response to the serious damage caused by chaotic pulse interference signals to the sorting and recognition of radar signals by the Electronic Intelligence System, a method based on Convolutional Neural Network—Bidirectional Long Short-Term Memory Network for suppressing cluttered pulse interference is proposed by introducing machine learning algorithms into the reconnaissance sorting process. This method utilizes Convolutional Neural Network to capture the locally distributed features of cluttered pulse interference signals across multiple dimensions, and Bidirectional Long Short-Term Memory network to capture the periodic features of radar signals within the mixed signals, thereby achieving classification and identification of interference signals and radar signals. Simulation experiments have shown that this method achieves an accuracy rate of over 96% and a recall rate of over 92% for different test sets. It has strong domain generalization ability and provides an effective way to improve the performance of ELINT systems in complex electromagnetic environments. |
| Key words: Electronic Intelligence System Convolutional Neural Network Bidirectional Long Short-Term Memory Network chaotic pulse interference radar reconnaissance |