摘要
近年来在实际生活中经常遇到各种飞行的无人机,而对于远距离的无人机存在着不能准确识别的情况,基于这一问题,根据每种无人机都具有独特的性质这一特点,使用一种基于无人机的微多普勒特征进行识别的AlexNet方法。首先使用77GHz的毫米波雷达对三种无人机进行探测,然后对雷达回波信号进行预处理,之后提取其微多普勒特征,最后通过AlexNet模型对特征数据集进行训练识别。结果显示,AlexNet模型对无人机种类的识别正确率达到93.75%。
In recent years,we have often encountered various flying drones in real life,and there are situations in which long-distance drones cannot be accurately identified.Based on this problem,each drone has unique properties.One feature is the use of an AlexNet method based on UAV's micro-Doppler feature for recognition.First,the 77GHz millimeter-wave radar is used to detect the three types of UAVs,and then the radar echo signals are preprocessed,and then the micro-Doppler features are extracted,and finally the feature data set is trained and recognized by the AlexNet model.The results show that the AlexNet model has a correct recognition rate of 93.75%for drone types.
作者
梁健涛
何法虎
LIANG Jiantao;HE Fahu(College of Artificial Intelligence,North China University of Science and Technology,Tangshan 063200)
出处
《现代计算机》
2021年第19期128-132,共5页
Modern Computer