基于神经网络的态势预测方法研究

为提高安全态势感知精确度和降低其复杂度,加强网络系统的安全防护能力,提出了基于神经网络的态势预测方法,并研究了基于径向基函数和遗传神经网络的态势动态预测方法。神经网络可通过自学习和非线性映射以任意精度逼近函数关系,具有高度灵活可变的拓扑结构及很强的适应能力,对于中、短期预测具有较高的精度。

Research on Situation Awareness Based on Neural Network

To improving the precision of security solution awareness and reducing its complexity,enforcing the defensive ability of network system,the method of situation awareness based on neural network was introduced,and the method of situation awareness based on Radial-Basis Function(RBF) and BP Neural Network with Genetic Algorithm(GABP) were researched.Though self-learning and nonlinear mapping,neural network can approach function relations at any precision.The method have the characteristics of flexible architecture,better adaptive capacity,and high accuracy in middle short term situation awareness.