一种稀缺样本条件下的雷达信号识别方法  

Towards Identifying Radar Signals with Sparse Data Samples

作　　者：于濛 杜天有[1] YU Meng;DU Tianyou

机构地区：[1]中国电子科技集团公司第二十研究所,西安710068

出　　处：《现代导航》2022年第4期303-309,共7页Modern Navigation

摘　　要：针对复杂电磁环境下的辐射源精确识别问题,基于战场环境下雷达信号采集困难、样本稀少的特点,以及相控阵雷达信号的特征,提出一种稀缺样本条件下基于迁移学习的雷达信号识别算法及其优化方法。该方法在识别稀缺样本目标时,训练时常短、识别准确度高,在-10 dB条件下,对四种雷达信号进行识别,识别准确率可达85.3%。Aiming at the problem of accurate identification of radiation sources in complex electromagnetic environments,a radar signal identification algorithm and its optimization method based on transfer learning are proposed under the condition of lacking samples.This method has the characteristics of short training time and high recognition rate when recognizing scare sample targets.Under the condition of-10 dB,four kinds of radar signals are recognized,reaching the recognition accuracy 85.3%.

关 键 词：深度学习 雷达信号 稀缺样本 迁移学习 

分 类 号：TN957[电子电信—信号与信息处理]