基于神经架构搜索及特征组合的便携式雷达个体识别方法  

A portable radar individual identification method based on neural architecture search and feature combination

作　　者：茆旋宇 王欢 王佩 秦晋 Mao Xuanyu;Wang Huan;Wang Pei;Qin Jin(No.8511 Research Institute of CASIC,Nanjing 210007,Jiangsu,China)

机构地区：[1]中国航天科工集团8511研究所,江苏南京210007

出　　处：《航天电子对抗》2024年第2期1-4,共4页Aerospace Electronic Warfare

摘　　要：电磁环境日趋复杂、雷达辐射源体制愈发多样,面向现代电子对抗需求的辐射源个体识别任务充满严峻挑战,特别是针对便携式雷达目标,个体识别模型性能会随着环境、参数变化、调制方式变换等因素出现显著下降。针对上述问题,提出了一种基于神经架构搜索及特征组合的个体识别方法,利用神经架构搜索自适应组建深度神经网络模型,对原始信号、不同特征之间的组合进行结构化调整,提高模型面对便携式雷达信号时的识别性能。The electromagnetic environment is becoming increasingly complex,and the radar radiation source system is becoming more diverse.The individual identification task of radiation sources facing modern elec-tronic countermeasures is full of severe challenges,especially for portable radar targets.The performance of indi-vidual identification models will significantly decline with changes in environment,parameters,modulation meth-ods,and other factors.In response to the above issues,an individual identification method based on neural archi-tecture search and feature combination is proposed.By using neural architecture search to adaptively construct a deep neural network model,the combination between the original signal and different features is structurally ad-justed to improve the model’s performance when confronted with portable radar signals.

关 键 词：神经架构搜索 个体识别 便携式雷达 

分 类 号：TN971.5[电子电信—信号与信息处理]