基于SOM-BP云神经网络的通信对抗作战能力评估  

Evaluation of Communication Countermeasure Combat Capability Based on SOM-BP Cloud Neural Networks

作　　者：史军涛[1] 周铭[2] 张振坤[3] 

机构地区：[1]空军预警学院,武汉430019 [2]武汉理工大学信息工程学院,武汉430019 [3]湖北青年职业学院,武汉430079

出　　处：《火力与指挥控制》2013年第10期99-102,共4页Fire Control & Command Control

摘　　要：为了增强神经网络训练样本的可靠性和正确率,提高网络自学习和自适应能力,使评估结果快速准确,针对通信对抗作战能力指标体系,在云神经网络的基础上,提出了基于SOM-BP云神经网络的效能评估方法。该方法利用SOM对样本数据进行聚类,选择具有特征性的样本对改进型的BP云神经网络进行训练。结果表明,该方法相对于BP神经网络,不仅收敛速度快,迭代次数少,而且准确率高。In order to enhance reliability and accuracy of neural network training samples,to improve self-learning and self-adapting ability of network,the evaluation result is made quickly and accurately. In this paper,according to the communication countermeasure combat capability index,on the basis of cloud-neural network,a evaluation method is presented base on SOM-BP cloud neural networks. This method uses SOM clustering for sample data,training the improved BP cloud-neural network by choosing the characteristic sample. The result shows that this method not only has faster convergence, fewer iterations, but also has the higher accuracy rate than BP cloud neural network.

关 键 词：自组织映像 BP云神经网络 通信对抗 效能评估 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] TN979[自动化与计算机技术—控制科学与工程]