基于布谷鸟搜索优化BP神经网络的网络安全态势评估方法

发布时间：2018-04-22 14:29

本文选题：态势评估 + 网络安全　；参考：《计算机应用》2017年07期

【摘要】：针对现有基于神经网络的网络安全态势评估方法效率低等问题,提出基于布谷鸟搜索(CS)优化反向传播(BP)神经网络(CSBPNN)的网络安全态势评估方法。首先,根据态势输入指标数和输出态势值确定BP神经网络(BPNN)的输入输出节点数,根据经验公式和试凑法计算出隐含层节点数;然后,随机初始化各层的连接权值和阈值,使用浮点数编码方式将权值与阈值编码成布谷鸟;最后,使用CS算法对权值和阈值进行优化,得到用于态势评估的CSBPNN模型并对其进行训练,将网络安全态势数据输入到CSBPNN模型中,获取网络的安全态势值。实验结果表明,与BPNN和遗传算法优化BP神经网络方法相比,基于CSBPNN的网络安全态势评估方法的迭代代数分别减少943和47且预测精度提高8.06个百分点和3.89个百分点,所提方法具有较快的收敛速度和较高的预测精度。
[Abstract]:Aiming at the low efficiency of the existing network security situation assessment methods based on neural network, a network security situation assessment method based on Cuckoo search (CSS) optimization and back propagation (BP) neural network (CSBPNN) is proposed. Firstly, the number of input and output nodes of BP neural network (BPNNN) is determined according to the number of input indicators and output situation values of BP neural network, and the number of hidden layer nodes is calculated according to empirical formula and trial and error method. Then, the connection weights and thresholds of each layer are initialized randomly. The weight and threshold are coded into cuckoo by floating point coding. Finally, the weight and threshold are optimized by CS algorithm, and the CSBPNN model for situation assessment is obtained and trained. The network security situation data is input into the CSBPNN model to obtain the network security situation value. The experimental results show that compared with the BP neural network method optimized by BPNN and genetic algorithm, the iterative algebra of the network security situation assessment method based on CSBPNN is reduced by 943 and 47 percent, respectively, and the prediction accuracy is increased by 8.06 percentage points and 3.89 percentage points, respectively. The proposed method has faster convergence rate and higher prediction accuracy.
【作者单位】：中国民航大学计算机科学与技术学院;
【基金】：国家科技重大专项(2012ZX03002002) 国家自然科学基金资助项目(60776807,61179045) 天津市科技计划重点项目(09JCZDJC16800) 中国民航科技基金资助项目(MHRD201009,MHRD201205)~~
【分类号】：TP18;TP393.08

【参考文献】