基于BP神经网络的网络安全态势预测研究

以网络系统安全运行作为出发点,从入侵检测系统和杀毒软件等安全手段中提取出可以反映网络安全运行的数据。从网络安全运行的威胁性、脆弱性、风险性和基础运行四个方面建立了一套网络安全态势评估体系。通过层次分析法对该评价体系进行计算,获得可以表征网络安全运行状态的综合安全指数。在该体系的基础上,利用BP神经网络对网络安全状态进行预测,实验结果表明,采用BP神经网络可以较好地对网络的安全态势进行预测,在相关研究领域有一定的借鉴价值。     

基于BP神经网络的温度预测方法

针对环境试验温度测量精度不高,测温延时较大的问题,提出了基于BP神经网络的温度预测方法。该方法通过测量瞬时温度变化率,利用BP神经网络的温度预测模型,对温度进行预测。在MATLAB中仿真表明,该方法对温度测量的精度有了明显的提高,对测温延时有了明显的改善。     

基于改进贝叶斯正则化BP神经网络模型的网络安全态势预测方法研究

随着互联网的迅速发展,网络安全问题越来越严重,分析及预测网络安网络安全态势,对于网络安全具有重要意义。本文在网络安全态势量化的的基础上,改进贝叶斯算法,提出一种改进型贝叶斯正则化BP神经网络模型的网络安全态势预测方法,通过模拟网络环境进行数据分析,验证了该预测方法可以减小了训练误差和预测误差,提高了对网络安全态势预测精度,证明了该方法的可行性。     

基于改进的递归神经网络的网络安全态势预测

本文利用神经网络处理非线性、复杂性等优势,基于改进的递归神经网络预测网络安全态势,实验结果证明该方法运行效率较高,运行结果与实际值相比,误差较低,精确性较高。     

基于改进的递归神经网络的网络安全态势预测

本文利用神经网络处理非线性、复杂性等优势,基于改进的递归神经网络预测网络安全态势,实验结果证明该方法运行效率较高,运行结果与实际值相比,误差较低,精确性较高。     

Quantitative method for network security situation based on attack prediction

To predict the attack behaviors accurately and comprehensively as well as to quantify the threat of attack,a quantitative method for network security situation based on attack prediction was proposed.By fusing the situation factors of attacker,defender and network environment,the capability of attacker and the exploitability rate of vulnerability were evaluated utilizing the real-time detected attack events,and the expected time-cost for attack-defense were further calculated.Then an attack prediction algorithm based on the dynamic Bayesian attack graph was designed to infer the follow-up attack actions.At last,the attack threat was quantified as the security risk situation from two levels of the hosts and the overall network.Experimental analysis indicates that the proposed method is suitable for the real adversarial network environment,and is able to predict the occurrence time of attack accurately and quantify the attack threat reasonably.     

Self-corrected coefficient smoothing method based network security situation prediction

In order to solve the problem of insufficient accuracy of current network security situation prediction methods,a new network security situation prediction model was proposed based on self-correcting coefficient smoothing.Firstly,a network security assessment quantification method was designed to transform the alarm information into situation real value time series based on the entropy correlation degree.Then,the adaptive solution of the static smoothing coefficient was calculated and the predicted initial value was obtained by using the variable domain space.Finally,based on the error category,the time-changing weighted Markov chain was built to modify the initial network situation prediction result and the prediction accuracy was further raised.The prediction model was tested with LL_DOS_1.0 dataset and the experimental results show that the proposed model has higher adaptability and prediction accuracy for network situation time series.     

A network security situation prediction model based on wavelet neural network with optimized parameters

The security incidents ion networks are sudden and uncertain, it is very hard to precisely predict the network security situation by traditional methods. In order to improve the prediction accuracy of the network security situation, we build a network security situation prediction model based on Wavelet Neural Network (WNN) with optimized parameters by the Improved Niche Genetic Algorithm (INGA). The proposed model adopts WNN which has strong nonlinear ability and fault-tolerance performance. Also, the parameters for WNN are optimized through the adaptive genetic algorithm (GA) so that WNN searches more effectively. Considering the problem that the adaptive GA converges slowly and easily turns to the premature problem, we introduce a novel niche technology with a dynamic fuzzy clustering and elimination mechanism to solve the premature convergence of the GA. Our final simulation results show that the proposed INGA-WNN prediction model is more reliable and effective, and it achieves faster convergence-speed and higher prediction accuracy than the Genetic Algorithm-Wavelet Neural Network (GA-WNN), Genetic Algorithm-Back Propagation Neural Network (GA-BPNN) and WNN.     

基于卡尔曼算法的网络安全态势预测方法分析

近几年,网络信息技术发展迅速,并逐渐进入社会各行各业的发展中,不仅改变着人们的生产方式,同时也改变了人们的生活方式,给人们衣食住行带来了很大的方便。但是应该看到网络在给人们带来极大便利的同时也存在较大的安全隐患,威胁人们的隐私和网络安全。所以必须要对网络安全态势进行分析和预测,找到一种合理的算法进行态势分析,建立科学的网络安全态势模型,为网络管理者做出安全防护的决策提供有效信息,提高网络安全度。这里主要利用卡尔曼算法进行网络安全态势预测方法的分析,并分析该算法的优越性。     

基于改进CRITIC的灰色关联网络安全态势评估方法

针对灰色关联分析算法无法定量评估样本、没有充分考虑到待评估指标间的关联性和决策者的期望问题,提出了一种基于改进的CRITIC(Criteria Importance Through Intercriteria Correlation)的灰色关联网络安全态势评估算法,借助于改进CRITIC的主客观赋权与TOPSIS(Te...     

Wi-Fi security evaluation model based on BP neural network

A Wi-Fi security evaluation model was proposed based on BP neural network.Firstly a large number of Wi-Fi hotspots was analyzed,safety information sources about wireless hotspots were selected.Secondly,a model of BP neural network was constructed,the weights were adjusted,and the evaluation system for Wi-Fi safety was formed,which was higher credibility.Finally,on the Android flatform the application based on above theory was designed.A lot of experi-ments show the model is able to scan and evaluate the wireless hotspot around user.It can also provide the disconnect function on unsafe wireless hotspot.     

基于灰色神经网络的车轮多边形阶次预测

针对于车轮多边形阶次发展趋势,运用高精度车轮周向不平顺检测设备进行测量,整合检测到的历史数据,提出了一种基于灰色神经网络的车轮多边形阶次预测模型。该模型充分运用了灰色模型计算量小,不需要大量样本数据的特点以及BP神经网络较强的非线性映射能力,利用该模型对车轮周向每个采样点的不平顺数据进行预测,再通过离散傅里叶变换和滤波处理得出该车轮各个阶次的粗糙度等级,以此来判断车轮多边形阶次的发展,较大程度上解决目前车轮镟修不及时等问题。     

基于BP神经网络技术的网络时延预测研究

为了应对网络控制系统中普遍存在的时延,有效预测是对网络控制系统非常重要的一环。介绍了一种基于BP神经网络的时延预测研究,利用BP神经网络中的最小平方算法来决定权值、基函数中心以及隐层的单元数量。运用Matlab软件对提出的BP神经网络的预测模型进行仿真实验,结果表明提出的BP神经网络模型在预测网络时延上非常有效。     

基于层次分析法和神经网络电网安全态势分析

对于电网的安全态势分析是依据评估指标完成的。由于安全评估指标间存在非线性关系,导致对安全态势分析的效果不佳。提出一种基于层次分析法和神经网络的电网安全态势分析评估方法,运用层次分析法对电网的多个逻辑结构进行分层,对影响安全态势的因素进行权值设定,获取各影响因素的权重,按照权重大小选择对电网安全干扰较高的影响因素,通过历史评估数据对神经网络模型进行训练后,完成电网的安全评估。实验结果说明改进方法可对电网安全进行准确评估,并且评估精度较高。     

基于BP神经网络模型的RSSI测距方法研究

针对传统信号传播路径损耗模型接收的信号强度指示(received signal strength indication, RSSI)测距误差较大, 提出了基于反向传播(back propagation, BP)神经网络模型的RSSI测距方法.首先, 研究分析传统信号传播路径损耗模型及测距误差; 其次, 利用BP神经网络构建新的路径损耗模型, 并将该模型应用到RSSI测距中, 对基于BP神经网络模型的RSSI测距方法进行研究; 最后, 通过实验和MATLAB仿真对测距方法进行验证.仿真结果表明:BP神经网络模型的RSSI测距误差比传统信号传播路径损耗模型的RSSI测距误差要小.     

基于BP神经网络的空调能耗预测与监控系统

针对空调能耗受室内外环境多因素影响而难以预测的问题,提出一种基于BP(Back Propagation)神经网络的空调能耗预测与监控系统。该方法将室外温湿度、房间冷热负荷、空调维持温湿度作为BP神经的输入,利用DeST分析并仿真在不同环境条件下的空调能耗。实验结果与样本数据比较可得,两者误差百分比在[-1.319%,1.270%]。为便于算法的实际应用,亦设计一套基于MVVM(Model-View-ViewModel)架构思想的空调能耗监测系统,并用Java将空调能耗预测方程进行封装。     

灰色关联分析与支持向量机相融合的网络安全态势评价

为了提高网络安全态势的评价准确性,提出一种灰色关联分析与支持向量机相融合的网络安全态势评价模型.首先采用灰色关联分析对网络安全态势评价指标进行筛选,并根据对评价结果贡献赋予评价指标权值,然后将重要的评价指标作为支持向量机的输入向量,并采用社会力模型算法选择模型的参数,最后采用仿真实例分析了模型的评价性能.实验结果表明,本文模型通过灰色关联分析选择支持向量机的输入向量和社会力模型算法选择了最合理的型参数,可以准确描述网络安全态势与评价指标之间的变化关系,不仅提高了网络安全态势评价的正确率,加快了建模速度,而且获得比经典模更优的评价结果.     

基于GANPSO-BP神经网络的粮情预测模型研究

在使用BP神经网络构建粮情温度预测模型时,因其存在误差高、稳定性差等缺陷,借鉴遗传算法和粒子群的思想,提出一种GANPSO-BP神经网络来预测粮食温度。首先为验证GANPSO算法的可用性,将该算法与PSO算法和IPSO算法在测试函数上利用Matlab软件进行模拟测试,结果得出GANPSO算法效果相对与其他两种算法有着明显提高;然后再对BP,PSO-BP和GANPSO-BP三种神经网络进行测试,得出BP的均方误差为0.021 79,PSO-BP的均方误差为0.017 65,GANPSO-BP的均方误差为0.013 30;从而得到GANPSO-BP神经网络相对于其他两种有着较好的稳定性,能够很好地预测粮食温度的变化情况。     

基于灰色理论和神经网络的房地产前期价格预测研究

为了克服单一灰色模型或者BP神经网络的局限性,提出灰色理论和神经网络的房地产前期价格预测模型。该模型采用灰色模型对房地产前期价格的规律性进行分析,BP神经网络对房地产前期价格的随机性进行刻画,实现了优势互补,最后进行房地产前期价格预测具体应用实例分析。结果表明,该模型可以准确描述房地产前期价格的变化趋势,提高房地产前期价格的预测精度,为房地产前期价格预测提供了一种有效的建模工具。