摘要
研究网络安全准确评估问题。网络安全评估是多指标、非线性、信息严重冗余的复杂信息传播过程,传统方法不能消除冗余信息,且不能准确描述非线性规律,网络安全评估结果的准确率低。为了提高网络安全评估的准确率,提出一种BP神经网络的网络安全评估模型。模型首先建立合理的网络安全评估指标体系,采用主成分析法对网络安全指标体系提取主要指标,消除指标间冗余息,最后将提取指标输入BP神经网络,通过BP神经网络的非线性学习能力对网络安全级别进行准确评估。在MATLAB平台上进行了仿真,结果表明,BP神经网络提高了网络安全评估结果的准确率,是一种高效、准确的网络安全评估方法,为保证网络安全提供了参考。
This paper discusses security assessment of computers,the traditional evaluation methods are affected by man＇s subjective arbitrariness,so the results are not ideal.In order to improve the accuracy of computer network security evaluation,we proposed an improved BP neural network computer network security evaluation model;the improved BP neural network used additional momentum method,variable step,and can overcome the disadvantages of slow convergence speed,easy falling in local optimum.Finally,.the results from MATLAB simulation experiment show that the improved BP neural network has effective recognition rate and higher precision,compared with the traditional BP neural network.The RMS error is only 0.013,which is much lower than traditional BP neural network.The result showed that the evaluation precision of improved BP neural network is much higher and the evaluation is more accurate.
出处
《计算机仿真》
CSCD
北大核心
2011年第6期177-180,共4页
Computer Simulation
关键词
网络安全
安全评估
人工神经网络
Network security
Security assessment
Artificial neural network
