流关联是一种链接网络流量以解决流身份鉴定的技术,但现有基于被动流量分析技术的流关联方式存在高存储和巨额计算开销问题。提出一种基于压缩感知—神经网络的流关联方法,该方法通过压缩感知中线性投影,对选取的流量特征进行降维处理,...流关联是一种链接网络流量以解决流身份鉴定的技术,但现有基于被动流量分析技术的流关联方式存在高存储和巨额计算开销问题。提出一种基于压缩感知—神经网络的流关联方法,该方法通过压缩感知中线性投影,对选取的流量特征进行降维处理,将降维处理后的流量特征作为卷积神经网络的输入,通过卷积神经网络进行关联特征提取,进而采用one class SVM分类器判断关联性。实验结果表明该方法在保持精确度的同时大大减少了存储、计算开销。展开更多
洋葱路由(Tor, the onion route)网络加密流的关联分析是其追踪溯源的核心技术之一;针对当前基于深度学习的流关联方法存在的时间特征不可靠且初级特征表征能力不强的问题,提出了一种基于时频分析和图卷积神经网络的关联分析方法,该方...洋葱路由(Tor, the onion route)网络加密流的关联分析是其追踪溯源的核心技术之一;针对当前基于深度学习的流关联方法存在的时间特征不可靠且初级特征表征能力不强的问题,提出了一种基于时频分析和图卷积神经网络的关联分析方法,该方法使用Tor网络流量的数据包长度信息作为原始特征序列,将数据包的包长度序列通过时频分布函数映射到时频域,并进一步将其嵌入为图结构数据,同时使用图卷积神经网络提取高阶特征,最后将得到的高阶特征输入三元组网络以实现相似流量的关联。实验结果表明误报率为0.1%时,所提方法的关联准确率可达到83.4%,明显优于已有的DeepCorr和Attcorr方法。展开更多
Different from the previous qualitative analysis of linear systems in time and frequency domains, the method for describing nonlinear systems quantitatively is proposed based on correlated dimensions. Nonlinear dynami...Different from the previous qualitative analysis of linear systems in time and frequency domains, the method for describing nonlinear systems quantitatively is proposed based on correlated dimensions. Nonlinear dynamics theory is used to analyze the pressure data of a contrarotating axial flow fan. The delay time is 18 and the embedded dimension varies from 1 to 25 through phase-space reconstruction. In addition, the correlated dimensions are calculated before and after stalling. The results show that the correlated dimensions drop from 1. 428 before stalling to 1. 198 after stalling, so they are sensitive to the stalling signal of the fan and can be used as a characteristic quantity for the judging of the fan stalling.展开更多
Based on the superposition principle of the nucleate boiling and convective heat transfer terms,a new correlation is developed for flow boiling heat transfer characteristics in helically coiled tubes.The effects of th...Based on the superposition principle of the nucleate boiling and convective heat transfer terms,a new correlation is developed for flow boiling heat transfer characteristics in helically coiled tubes.The effects of the geometric and system parameters on heat transfer characteristics in helically coiled tubes are investigated by collecting large amounts of experimental data and analyzing the heat transfer mechanisms. The existing correlations are divided into two categories,and they are calculated with the experimental data.The Dn factor is introduced to take into account the effect of a complex geometrical structure on flow boiling heat transfer.A new correlation is developed for predicting the flow boiling heat transfer coefficients in the helically coiled tubes,which is validated by the experimental data of R134a flow boiling heat transfer in them;and the average relative error and root mean square error of the new correlation are calculated.The results show that the new correlation agrees well with the experimental data,indicating that the new correlation can be used for predicting flow boiling heat transfer characteristics in the helically coiled tubes.展开更多
文摘流关联是一种链接网络流量以解决流身份鉴定的技术,但现有基于被动流量分析技术的流关联方式存在高存储和巨额计算开销问题。提出一种基于压缩感知—神经网络的流关联方法,该方法通过压缩感知中线性投影,对选取的流量特征进行降维处理,将降维处理后的流量特征作为卷积神经网络的输入,通过卷积神经网络进行关联特征提取,进而采用one class SVM分类器判断关联性。实验结果表明该方法在保持精确度的同时大大减少了存储、计算开销。
文摘洋葱路由(Tor, the onion route)网络加密流的关联分析是其追踪溯源的核心技术之一;针对当前基于深度学习的流关联方法存在的时间特征不可靠且初级特征表征能力不强的问题,提出了一种基于时频分析和图卷积神经网络的关联分析方法,该方法使用Tor网络流量的数据包长度信息作为原始特征序列,将数据包的包长度序列通过时频分布函数映射到时频域,并进一步将其嵌入为图结构数据,同时使用图卷积神经网络提取高阶特征,最后将得到的高阶特征输入三元组网络以实现相似流量的关联。实验结果表明误报率为0.1%时,所提方法的关联准确率可达到83.4%,明显优于已有的DeepCorr和Attcorr方法。
基金Supported by the Natural Science Foundation of Jiangsu Province(BK2005018)the Graduate Research and Innovation Plan of Jiangsu Province(CX07B-061Z)~~
文摘Different from the previous qualitative analysis of linear systems in time and frequency domains, the method for describing nonlinear systems quantitatively is proposed based on correlated dimensions. Nonlinear dynamics theory is used to analyze the pressure data of a contrarotating axial flow fan. The delay time is 18 and the embedded dimension varies from 1 to 25 through phase-space reconstruction. In addition, the correlated dimensions are calculated before and after stalling. The results show that the correlated dimensions drop from 1. 428 before stalling to 1. 198 after stalling, so they are sensitive to the stalling signal of the fan and can be used as a characteristic quantity for the judging of the fan stalling.
基金The National Natural Science Foundation of China(No.50776055,51076084)
文摘Based on the superposition principle of the nucleate boiling and convective heat transfer terms,a new correlation is developed for flow boiling heat transfer characteristics in helically coiled tubes.The effects of the geometric and system parameters on heat transfer characteristics in helically coiled tubes are investigated by collecting large amounts of experimental data and analyzing the heat transfer mechanisms. The existing correlations are divided into two categories,and they are calculated with the experimental data.The Dn factor is introduced to take into account the effect of a complex geometrical structure on flow boiling heat transfer.A new correlation is developed for predicting the flow boiling heat transfer coefficients in the helically coiled tubes,which is validated by the experimental data of R134a flow boiling heat transfer in them;and the average relative error and root mean square error of the new correlation are calculated.The results show that the new correlation agrees well with the experimental data,indicating that the new correlation can be used for predicting flow boiling heat transfer characteristics in the helically coiled tubes.