以Tor网络为代表的匿名网络在带来强隐私性保护的同时也为网络违法犯罪活动提供了温床,因此,开展实时、高精度的Tor网络流量识别研究具有重要的现实意义。为此,针对现有研究存在泛化性不强和实时性差等问题,提出了一种基于多模态特征融...以Tor网络为代表的匿名网络在带来强隐私性保护的同时也为网络违法犯罪活动提供了温床,因此,开展实时、高精度的Tor网络流量识别研究具有重要的现实意义。为此,针对现有研究存在泛化性不强和实时性差等问题,提出了一种基于多模态特征融合和Stacking集成学习技术的Tor网络流量识别方法rtTorTIM。具体来讲,该方法首先提取Tor网络流量的主机级、流级和包级3种模态相关特征并构造特征数据集;随后,rtTorTIM选取随机森林、线性回归和K-近邻方法作为基学习器,并使用一个线性神经网络进行决策融合,从而构建起一个2层Stacking流量分类器。基于ISCX Tor 2016公开数据集的对比实验结果表明,rtTorTIM方法在Tor流量识别上的准确率、精确率和召回率均达到了99%,同时该方法在分类实时性上也展现出更优的性能。展开更多
Emerging two-dimensional MXenes have been extensively studied in a wide range of fields thanks to their superior electrical and hydrophilic attributes as well as excellent chemical stability and mechanical flexibility...Emerging two-dimensional MXenes have been extensively studied in a wide range of fields thanks to their superior electrical and hydrophilic attributes as well as excellent chemical stability and mechanical flexibility.Among them,the ultrahigh electrical conductivity(σ)and tunable band structures of benchmark Ti_(3)C_(2)T_(x) MXene demonstrate its good potential as thermoelectric(TE)materials.However,both the large variation ofσreported in the literature and the intrinsically low Seebeck coefficient(S)hinder the practical applications.Herein,this study has for the first time systematically investigated the TE properties of neat Ti_(3)C_(2)T_(x) films,which are finely modulated by exploiting different dispersing solvents,controlling nanosheet sizes and constructing composites.First,deionized water is found to be superior for obtaining closely packed MXene sheets relative to other polar solvents.Second,a simultaneous increase in both S andσis realized via elevating centrifugal speed on MXene aqueous suspensions to obtain small-sized nanosheets,thus yielding an ultrahigh power factor up to~156μW m^(-1) K^(-2).Third,S is significantly enhanced yet accompanied by a reduction inσwhen constructing MXene-based nanocomposites,the latter of which is originated from the damage to the intimate stackings of MXene nanosheets.Together,a correlation between the TE properties of neat Ti_(3)C_(2)T_(x) films and the stacking of nanosheets is elucidated,which would stimulate further exploration of MXene TEs.展开更多
The adoption of deep learning-based side-channel analysis(DL-SCA)is crucial for leak detection in secure products.Many previous studies have applied this method to break targets protected with countermeasures.Despite ...The adoption of deep learning-based side-channel analysis(DL-SCA)is crucial for leak detection in secure products.Many previous studies have applied this method to break targets protected with countermeasures.Despite the increasing number of studies,the problem of model overfitting.Recent research mainly focuses on exploring hyperparameters and network architectures,while offering limited insights into the effects of external factors on side-channel attacks,such as the number and type of models.This paper proposes a Side-channel Analysis method based on a Stacking ensemble,called Stacking-SCA.In our method,multiple models are deeply integrated.Through the extended application of base models and the meta-model,Stacking-SCA effectively improves the output class probabilities of the model,leading to better generalization.Furthermore,this method shows that the attack performance is sensitive to changes in the number of models.Next,five independent subsets are extracted from the original ASCAD database as multi-segment datasets,which are mutually independent.This method shows how these subsets are used as inputs for Stacking-SCA to enhance its attack convergence.The experimental results show that Stacking-SCA outperforms the current state-of-the-art results on several considered datasets,significantly reducing the number of attack traces required to achieve a guessing entropy of 1.Additionally,different hyperparameter sizes are adjusted to further validate the robustness of the method.展开更多
文摘以Tor网络为代表的匿名网络在带来强隐私性保护的同时也为网络违法犯罪活动提供了温床,因此,开展实时、高精度的Tor网络流量识别研究具有重要的现实意义。为此,针对现有研究存在泛化性不强和实时性差等问题,提出了一种基于多模态特征融合和Stacking集成学习技术的Tor网络流量识别方法rtTorTIM。具体来讲,该方法首先提取Tor网络流量的主机级、流级和包级3种模态相关特征并构造特征数据集;随后,rtTorTIM选取随机森林、线性回归和K-近邻方法作为基学习器,并使用一个线性神经网络进行决策融合,从而构建起一个2层Stacking流量分类器。基于ISCX Tor 2016公开数据集的对比实验结果表明,rtTorTIM方法在Tor流量识别上的准确率、精确率和召回率均达到了99%,同时该方法在分类实时性上也展现出更优的性能。
基金supported by the China Postdoctoral Science Foundation(grant No.2024M750511,J.T.)National Key R&D Program of China(grant No.2022YFB3603804,Y.Z.)National Natural Science Foundation of China(NSFC)under grant Nos.82172470(C.X.)and 22375050(Z.L.).
文摘Emerging two-dimensional MXenes have been extensively studied in a wide range of fields thanks to their superior electrical and hydrophilic attributes as well as excellent chemical stability and mechanical flexibility.Among them,the ultrahigh electrical conductivity(σ)and tunable band structures of benchmark Ti_(3)C_(2)T_(x) MXene demonstrate its good potential as thermoelectric(TE)materials.However,both the large variation ofσreported in the literature and the intrinsically low Seebeck coefficient(S)hinder the practical applications.Herein,this study has for the first time systematically investigated the TE properties of neat Ti_(3)C_(2)T_(x) films,which are finely modulated by exploiting different dispersing solvents,controlling nanosheet sizes and constructing composites.First,deionized water is found to be superior for obtaining closely packed MXene sheets relative to other polar solvents.Second,a simultaneous increase in both S andσis realized via elevating centrifugal speed on MXene aqueous suspensions to obtain small-sized nanosheets,thus yielding an ultrahigh power factor up to~156μW m^(-1) K^(-2).Third,S is significantly enhanced yet accompanied by a reduction inσwhen constructing MXene-based nanocomposites,the latter of which is originated from the damage to the intimate stackings of MXene nanosheets.Together,a correlation between the TE properties of neat Ti_(3)C_(2)T_(x) films and the stacking of nanosheets is elucidated,which would stimulate further exploration of MXene TEs.
基金supported by the Hunan Provincial Natural Science Foundation of China(2022JJ30103)“the 14th Five-Year Plan”Key Disciplines and Application-Oriented Special Disciplines of Hunan Province(Xiangjiaotong[2022]351)the Science and Technology Innovation Program of Hunan Province(2016TP1020).
文摘The adoption of deep learning-based side-channel analysis(DL-SCA)is crucial for leak detection in secure products.Many previous studies have applied this method to break targets protected with countermeasures.Despite the increasing number of studies,the problem of model overfitting.Recent research mainly focuses on exploring hyperparameters and network architectures,while offering limited insights into the effects of external factors on side-channel attacks,such as the number and type of models.This paper proposes a Side-channel Analysis method based on a Stacking ensemble,called Stacking-SCA.In our method,multiple models are deeply integrated.Through the extended application of base models and the meta-model,Stacking-SCA effectively improves the output class probabilities of the model,leading to better generalization.Furthermore,this method shows that the attack performance is sensitive to changes in the number of models.Next,five independent subsets are extracted from the original ASCAD database as multi-segment datasets,which are mutually independent.This method shows how these subsets are used as inputs for Stacking-SCA to enhance its attack convergence.The experimental results show that Stacking-SCA outperforms the current state-of-the-art results on several considered datasets,significantly reducing the number of attack traces required to achieve a guessing entropy of 1.Additionally,different hyperparameter sizes are adjusted to further validate the robustness of the method.
