采用认知雷达架构有助于实现雷达抗干扰技术的智能化程度提升。针对认知抗干扰技术领域中雷达对电磁干扰环境的感知问题,文中提出了一种基于深度学习的多节点干扰调制类型识别方法。该方法针对雷达信号处理的不同节点,如数字波束形成、自适应副瓣对消、脉冲压缩前后以及动目标检测之后,采用多个节点的时频平面和距离多普勒平面作为干扰信号的联合特征提取对象,建立了基于深度学习的多节点干扰识别策略模型,以提高多种干扰场景下的干扰识别正确率。为了提升干扰特征的提取能力和网络的训练效率,用于干扰识别的深度学习算法在卷积神经网络(CNN)的基础上引入了注意力机制和残差网络,建立了针对多节点策略下的干扰类型识别网络结构,实现了对多种不同干扰场景下的干扰类型识别。仿真结果表明,在单一干扰场景下,当干噪比为14 dB 时,所提算法的干扰识别准确率可达92%。在多干扰场景下,所提算法在不同节点策略的加持下,识别准确率可达90%。

In the synergy application of detection group for netted radar system, it has been difficult for commanders to quickly and correctly decide on group-building forms and control plans, which has led to difficulties in agile group-building and precise control, restricting the synergy detection efficiency of new air-space threats. On the basis of briefly describing the principle and complexity of human-computer decision fusion, this paper reveals the mechanism of human-computer decision fusion technology in generating synergy detection complexity and cracking the complexity of new air-space threats. It establishes the model of human-computer fused decision

TN957

“十四五” 全军共用信息系统装备预先研究项目(315027502)