基于离散Hopfield神经网络的辐射源目标威胁估计  

Threat Assessment for Radiation Source Target Based on Discrete Hopfield Neural Network

作　　者：张永利 潘哲 刘楠楠 ZHANG Yongli;PAN Zhe;LIU Nannan(China Academy of Electronics and Information Technology of CETC,Beijing 100041,China;The Fourth Military Representative Office of Air Force Equipment Department in Beijing,Beijing 100041,China)

机构地区：[1]中国电子科技集团公司电子科学研究院,北京100041 [2]空装驻北京地区第四军事代表室,北京100041

出　　处：《指挥信息系统与技术》2020年第2期39-43,48,共6页Command Information System and Technology

基　　金：装备发展部“十三五”预研课题资助项目。

摘　　要：利用离散Hopfield神经网络对获得的历史情报数据进行挖掘,形成理想的威胁评估指标体系,将待分类的威胁目标评价指标输入创建的离散Hopfield神经网络,实现对辐射源目标的威胁等级评估。将基于离散Hopfield神经网络的辐射源目标评估等级与基于灰色关联投影法获得的威胁评估等级进行比较,其评估结果一致。离散Hopfield神经网络具有自组织自适应性、并行处理和高容错性等特性,适用于解决数据发掘问题。The discrete Hopfield neural network is used to mine the acquired historical information data to form an ideal threat assessment index system.The threat target index system to be classified is input into the created discrete Hopfield neural network to realize threat level evaluation of radiation source targets.The radiation source evaluation level obtained based on the discrete Hopfield neural network is compared with that based on the gray relational projection method,and the evaluation results are consistent.The discrete Hopfield neural network has the characteristics of self-organizing,self-adaptation,parallel processing,and high fault tolerance,therefore it is suitable for solving data mining problems.

关 键 词：离散HOPFIELD神经网络 威胁估计 威胁等级 

分 类 号：TN918[电子电信—通信与信息系统]