| 摘要: |
| 开展脉冲重复间隔（Pulse Repetition Interval,PRI）模式识别工作是电子支援系统的一项重要任务。现代复杂电磁环境下，受雷达辐射源部署和接收设备本身影响，雷达脉冲丢失率极高，导致分选后PRI序列调制规律被破坏，现有的PRI模式识别方法准确率不足。针对上述问题，从PRI序列还原角度出发，并结合PRI序列本质是时序序列的特点，提出GAIN-LSTM（Generative Adversarial Imputation Nets and Long Short Term Memory）网络架构，其先对丢失脉冲位置进行补全操作，恢复PRI调制规律，然后对还原后PRI序列进行调制模式识别。仿真结果表明，提出的GAIN-LSTM网络架构在脉冲丢失率70%时仍保持95%的正确识别率。 |
| 关键词: 脉冲重复间隔调制 数据补全 模式识别 GAIN-LSTM |
| DOI：10.20079/j.issn.1001-893x.230106001 |
|
| 基金项目:四川省重点研发计划项目(2020YFG0051)；校企合作项目(21H1445) |
|
| A Radar PRI Sequence Restoration and Recognition Method Based on GAIN-LSTM Network |
| LI Zhongyuan,XIAN Guo,GONG Xiaofeng,LUO Ruisen |
| (1.College of Electrical Engineering,Sichuan University,Chengdu 610065,China;2.Chengdu Dagongbochuang Information Technology Co.,Ltd.,Chengdu 610065,China) |
| Abstract: |
| For an electronic support system,it is essential to recognize the modulation of pulse repetition intervals(PRIs),since it is directly related to the indication of radar emitters.However,PRI modulations are more difficult to recognize due to the high ratio of lost pulses.From the perspective of PRI sequence restoration,and according to the characteristics that PRI sequence is essentially a time sequence,the authors propose a novel method named Generative Adversarial Imputation Nets and Long Short Term Memory(GAIN-LSTM) network architecture.This method first completes the missing pulse position to restore the PRI modulation rule,and then carries out the modulation pattern recognition.The simulation results show that the proposed GAIN-LSTM network architecture maintains 95% |
| Key words: pulse repetition interval modulation data reconstruction pattern recognition GAIN-LSTM network |
