基于深度学习的ADS-B辐射源个体识别

辐射源个体识别技术在电子对抗领域起着关键性作用.针对传统人工提取特征主要依赖于专家经验,文中提出了一种基于深度学习的ADS-B辐射源个体识别方法.该方法采用卷积神经网络结合center loss损失函数来训练和测试相位数据.并加入假冒飞机序列号,查看中心损失值来判断是否异常.实验表明,文中提出的方法识别精度能达到99％...

ADS-B signal radiation source recognition based on deep learning

The individual identification technology of radiation sources plays a key role in the field of electronic warfare.Aiming at the traditional manual feature extraction mainly relying on expert experience,this paper proposes an ADS-B radiation source individual identification method based on deep learning.This method uses convolutional neural network combined with center loss function to train and test phase data.And add the fake aircraft serial number,check the center loss value to judge whether it is abnormal.The experimental results show that the recognition accuracy of the method proposed in this paper can reach about 99%,and it can distinguish real messages from fake messages.