基于LVQ神经网络的雷达杂波抑制方法  

Radar Clutter Suppression Method Based on LVQNeural Network

作　　者：施端阳 林强[1] 胡冰[1] 翟芸 SHI Duanyang;LIN Qiang;HU Bing;ZHAI Yun(Air-Defense Early Warning Equipment Department,Air Force Early Warning Academy,Wuhan 430019,China;Unit 95174 of PLA,Wuhan 430040,China)

机构地区：[1]空军预警学院防空预警装备系,武汉430019 [2]解放军95174部队,武汉430040

出　　处：《火力与指挥控制》2023年第4期37-44,共8页Fire Control & Command Control

基　　金：全军军事类研究生重点基金资助项目(JY2020B150)。

摘　　要：针对雷达目标检测后的剩余杂波影响雷达航迹起始和航迹跟踪的问题,提出基于学习向量量化(learning vector quantization,LVQ)神经网络的雷达杂波抑制方法。从雷达回波点迹的特征入手,通过分析目标点迹和杂波点迹的特征分布,通过人工提取特征的方式选取具有差异化的特征。根据特征数量和点迹类别数量构建LVQ神经网络分类模型,并对模型进行训练。利用训练好的LVQ神经网络分类模型对雷达回波点迹进行分类,区分目标点迹和杂波点迹,保留判别为目标的点迹,滤除判别为杂波的点迹,从而实现杂波抑制功能。通过对某型航管雷达的实测数据进行测试表明:该方法能够有效区分目标点迹和杂波点迹,杂波抑制能力比BP神经网络算法更好。Aiming at the problem that the residual clutter after radar target detection affects the initiation of radar track and the tracking of flight path,a radar clutter suppression method based on learning vector quantization(LVQ)neural network is proposed.Firstly,starting with the characteristics of radar echo trace points,by analyzing the feature distribution of target trace points and clutter trace points,the differentiated features are selected by manual feature extraction.Then,the trained LVQ neural network classification model is constructed according to the number of features and the number of trace points categories,and the model is trained.Finally,the trained LVQ neural network classification model is used to classify the radar echo points and to distinguish the target trace points and clutter points,and to retain the trace points discriminated as the targets and filter out the points discriminated as clutter so as to realize the clutter suppression function.By testing the measured data of an ATC radar,it is found that this method can effectively distinguish target trace points and clutter trace points,and the clutter suppression ability is better than that of BP neural network algorithm.

关 键 词：雷达 杂波抑制 LVQ神经网络 点迹处理 

分 类 号：TN957[电子电信—信号与信息处理]