基于CNN和D-S证据理论的多站协同多功能雷达工作模式识别方法

A Recognition Method for Working Pattern of Multi-Station Cooperative Multi-Function Radar Based on CNN and D-S Evidence Theory

传统的多功能雷达工作模式识别方法主要利用单一电子侦察设备侦收的脉冲数据完成特征提取,其模型或算法的泛化能力不强。因此,提出基于卷积神经网络(Convolutional Neural Network,CNN)和D-S(Dempster-Shafer)证据理论的多站协同多功能雷达工作模式识别方法。首先,利用轻量级CNN模型对不同方位、不同俯仰下侦察截获分选的脉冲幅度/波形单元数据进行自适应特征提取。其次,利用多站协同侦察系统通过D-S证据理论融合多站协同下不同侦察站点的Softmax分类器的分类结果,实现在差侦察条件下对不同空间方向上的多功能雷达工作模式快速准确识别。仿真验证结果表明,该方法相比于单侦察站条件下具有更好的识别性能。

The traditional multi-function radar working mode recognition method mainly uses pulse data detected by a single electronic reconnaissance equipment to complete the feature ex-traction.The generalization ability of the model or algorithm is not strong.Therefore,a multi-sta-tion cooperative multi-function radar working pattern recognition method based on convolutional neural network(CNN)and dempster-shafer(D-S)evidence theory is proposed.Firstly,the lightweight CNN model is used to perform adaptive feature extraction on the pulse amplitude/waveform unit data of reconnaissance interception and sorting in different azimuths and different pitches.The classification results of the Softmax classifier at the reconnaissance site can quickly and accurately identify the working modes of multi-functional radars in different spatial direc-tions under poor reconnaissance conditions.The simulation results show that the method has better generalization ability than that of a single reconnaissance station.