基于GAN的雷达HRRP数据增强方法  

Radar HRRP Data Enhancement Method Based on GAN

作　　者：周强 王彦华[1,3] 宋益恒 李阳 ZHOU Qiang;WANG Yanhua;SONG Yiheng;LI Yang(Radar Research Laboratory,School of Information and Electronics,Beijing Institute of Technology,Beijing 100081,China;Beijing Key Laboratory of Embedded Real-time Information Processing Technology,Beijing 100081,China;Beijing Institute of Technology Chongqing Innovation Center,Chongqing 401120,China)

机构地区：[1]北京理工大学信息与电子学院雷达技术研究所,北京100081 [2]嵌入式实时信息处理技术北京市重点实验室,北京100081 [3]北京理工大学重庆创新中心,重庆401120

出　　处：《信号处理》2022年第1期92-99,共8页Journal of Signal Processing

基　　金：国家重点研发计划(2018YFE0202101,2018YFE0202102,2018YFE0202103);国家自然科学基金(61701026)。

摘　　要：在雷达自动目标识别(RATR)中,数据驱动方法是强有力的工具之一。然而数据驱动方法的性能十分依赖数据集的质量,数据增强方法通过扩充数据集,能够提升数据驱动模型在现有数据集上的识别率。本文提出了用于高分辨距离像(HRRP)数据生成的一维基础生成对抗网络(BGAN)结构和条件生成对抗网络(CGAN)结构,并利用生成的人工样本补充不完备数据集完成了数据增强。实验表明,本文所提出的两种网络均能有效提升目标识别的准确率,提升效果优于传统的平移和镜像增强方法。基于BGAN的HRRP数据增强方法提升效果最优,但其模型时间与空间复杂度较高;基于CGAN的数据增强方法能够在保证识别率提升的同时降低模型的时间与空间复杂度,具有较高的应用前景。In Radar Automatic Target Recognition(RATR),data-driven models have proven to be a powerful tool.However,the performance of the data-driven models were dependent on the quality of the data set.The data enhancement method could improve the recognition performance of the data-driven models on the existing data set by expanding the data set.This paper proposes a one-dimensional basic generative adversarial network(BGAN)structure and a conditional generative adversarial network(CGAN)structure for high resolution range profile(HRRP)data generation.Then using the generated artificial samples to complete the data enhancement.Experiments show that the two networks proposed in this paper can effectively improve the accuracy of target recognition,and the performance is better than the traditional translation and mirroring enhancement methods.The BGAN-based HRRP data enhancement method has the best performance,but its time and space complexity are relatively high;the CGAN-based data enhancement method can reduce the time and space complexity of the model while ensuring the increase in accuracy,and has high application prospects.

关 键 词：雷达目标识别 高分辨距离像 数据增强 生成对抗网络 

分 类 号：TN959.1[电子电信—信号与信息处理]