检索结果-内蒙古大学图书馆

GSA-DT: A Malicious Traffic Detection Model Based on Graph Self-Attention Network and Decision Tree

IEEE Transactions on Network and Service Management 2025年第2期22卷 2059-2073页

作者： Cai, Saihua Tang, Han Chen, Jinfu Lv, Tianxiang Zhao, Wenjun Huang, Chunlei Jiangsu University School of Computer Science and Communication Engineering Zhenjiang212013 China Jiangsu Provincial Key Laboratory of Industrial Cyberspace Security Technology Zhenjiang212013 China

Malicious attack has shown a rapid growth in recent years, it is very important to accurately detect malicious traffic to defend against malicious attacks. Compared with machine learning and deep learning technologies, graph convolutional neural network (GCN) achieves better detection results of malicious traffic due to additional consideration of the correlation between network traffic features. However, existing GCN-based detection models suffer from fixed weight assignment, only focusing on local features, lack the ability to model graph structure and relationships as well as having gradient disappearance. To solve these problems, this paper proposes the GSA-DT model based on graph self-attention network and decision tree. GSA-DT first preprocesses the original network traffic to obtain better traffic features and labels, and then uses GCN to extract the topological structure of network traffic as well as capture the correlation relationships among traffic features, where the ReLU activation function is replaced by LeakyReLU to overcome the problems of neuron "death"and gradient disappearance during the training process;It also introduces the self-attention mechanism into GCN to assign larger weights to the key features to reduce the interference of redundant features. Finally, GSA-DT uses decision tree to perform the detection of malicious traffic. Experimental results on five network traffic datasets show that GSA-DT model improves the detection accuracy over 1% on average than seven advanced malicious traffic detection models, and it also performs better in F1-measure, TPR, FPR as well as stability. © 2004-2012 IEEE.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Analysis of security for a New Kind of an Extended Generalized Feistel Network 24

Analysis of Security for a New Kind of an Extended Generaliz...

引用

3rd International Conference on Algorithms, Data Mining, and Information technology, ADMIT 2024

作者： Dong, Xinfeng Wu, You Zhou, Yu Tan, Hao Zhang, Wenzheng University of Electronic Science and Technology of China Sichuan Chengdu China National Key Laboratory of Security Communication Chengdu China

ISBN: (纸本)9798400718120

Differential cryptanalysis and linear cryptanalysis are two classic analytical methods for block ciphers, and they are of great significance for evaluating the security of block ciphers. In practical security evaluations, a commonly used approach is to study lower bounds on the number of active F-functions in multi-round differential characteristics and linear approximations of the cipher structure, thereby establishing upper bounds on the maximum differential characteristic probability and the maximum linear approximation probability. If these upper bounds are sufficiently small, the cipher structure is considered to have resistance against differential and linear cryptanalysis. Based on this, this paper conducts a detailed study on the resistance of a class of four-branch extended generalized Feistel networks against differential and linear cryptanalysis. Under the condition that the F-function is bijective, it is proven that the lower bound on the number of active F-functions in k-round (k ≥ 1) differential characteristics or linear approximations is k − 1. Thus, if the maximum differential probability and the maximum linear approximation probability of the F-function are denoted as p and q, respectively, the upper bounds on the k-round differential characteristic probability and linear approximation probability are pk−1 and qk−1, respectively. © 2024 Copyright held by the owner/author(s).

关键词： Ciphertext

来源：评论

学校读者我要写书评

暂无评论

Enhanced Data Mining Algorithms for Sybil Attack Prevention in Truth Discovery 24

Enhanced Data Mining Algorithms for Sybil Attack Prevention ...

引用

3rd International Conference on Algorithms, Data Mining, and Information technology, ADMIT 2024

作者： Shao, Yanjing Zhao, Wei Liu, Jianghua Xu, Lei Xu, Chungen School of Mathematics and Statistics Nanjing University of Science and Technology China National Key Laboratory of Security Communication China School of Computer Science and Engineering Nanjing University of Science and Technology China

ISBN: (纸本)9798400718120

In crowdsourcing systems, where substantial amounts of data from various contributors are aggregated to discern reliable information, privacy concerns are often managed through differential privacy techniques. However, these systems are vulnerable to Sybil attacks, where attackers create multiple fake contributors to gain undeserved rewards. The presence of differential privacy noise complicates the detection of these fraudulent contributors, thereby reducing the accuracy of truth discovery. To address this challenge, we introduce an enhanced data mining algorithm, Truth Discovery against Sybil Attacks (TDSA). This algorithm is specifically designed to improve truth discovery by effectively mitigating Sybil attacks and reducing transmission errors in data mining contexts, utilizing machine learning clustering algorithms and lattice-based linear coding techniques. TDSA optimizes the mining process to better handle malicious data contributions and reduces errors associated with data aggregation. Experimental results on real-world datasets demonstrate that TDSA improves accuracy by approximately 20% compared to conventional methods lacking Sybil attack defenses, highlighting its efficacy in advanced data mining applications. © 2024 Copyright held by the owner/author(s).

关键词： Data aggregation

来源：评论

学校读者我要写书评

暂无评论

Anti-concept Drift-enabled Recognition Method of Encrypted Traffic Behavior 24

Anti-concept Drift-enabled Recognition Method of Encrypted T...

引用

2nd International Conference on communication Networks and Machine Learning, CNML 2024

作者： Zheng, Jicheng Huang, Shuangshuang Ji, Qingbing School of Computer Science and Technology Xi’an Jiaotong University Xi’an China National Key Laboratory of Security Communication Sichuan Chengdu China

ISBN: (纸本)9798400711688

Although encrypted traffic ensures information security, it can also be exploited by malicious actors, leading to security risks. This paper addresses the limitations of existing behavior recognition technologies, particularly the concept drift issue in the"training-testing" phase. We propose an Anti-Concept Drift Framework (ACD-WFP), which includes a Confusion-Aware Adversarial Concept Drift Detection Algorithm (CACD) and an Incremental Learning-Based Multi-Modal Behavior Recognition Algorithm (IM-WFP). In experiments involving 75GB of encrypted traffic across 12 concept drift scenarios, CACD efficiently detects and tracks drift with low storage requirements, while IM-WFP improves recognition accuracy by 2% to 22% through continuous learning of new concepts. © 2024 Copyright held by the owner/author(s).

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

DialTest-EA: An Enhanced Fuzzing Approach With Energy Adjustment for Dialogue Systems via Metamorphic Testing

DialTest-EA: An Enhanced Fuzzing Approach With Energy Adjust...

引用

作者： Chen, Haibo Chen, Jinfu Wu, Yucheng Cai, Saihua Ahmad, Bilal Huang, Rubing Wang, Shengran Zhang, Chi School of Computer Science and Communication Engineering Jiangsu Key Laboratory of Security Technology Industrial Cyberspace Jiangsu University Zhenjiang China School of Computer Science and Engineering Faculty of Innovation Engineering Macau University of Science and Technology China

Deep neural networks (DNNs) possess potent feature learning capability, enabling them to comprehend natural language, which strongly support developing dialogue systems. However, dialogue systems usually perform incorrect behaviours in some corner cases, which may cause misunderstanding or economic loss. To test and debug dialogue systems, a popular fuzzing framework by metamorphic testing with Gini impurity guidance is proposed, namely, DialTest. However, DialTest treats all seeds (the initial test inputs to generate the mutated test inputs) equally during the fuzzing process and does not differentiate seeds, resulting in a certain limitation to its incorrect behaviour detection capability. In this paper, we propose to enhance the DialTest by applying a lightweight energy adjustment strategy called DialTest with Energy Adjustment (DialTest-EA). DialTest-EA employs the ant colony optimization algorithm (ACO) to adjust the mutation energy of each seed adaptively, ensuring that potential seeds have more opportunities to generate subsequent test inputs. To evaluate the effectiveness of the proposed DialTest-EA, we conduct a series of comparisons with the original DialTest and random mutation strategy. The experimental results show that the proposed DialTest-EA outperforms the compared methods both in the intent detection and slot filling tasks. Compared with the original DialTest, the intent detection accuracy of generated test cases by the proposed method is reduced by more than 14%, and the slot filling accuracy is reduced by more than 8%. © 2024 John Wiley & Sons Ltd.

关键词： Ant colony optimization

来源：评论

学校读者我要写书评

暂无评论

Generative Replay and Multi-steps Knowledge Distillation in Class-Incremental Learning 24

Generative Replay and Multi-steps Knowledge Distillation in ...

引用

3rd International Conference on Artificial Intelligence and Intelligent Information Processing, AIIIP 2024

作者： Meng, Yicheng Ping, Jinming Shi, Jingye Zhi, Ruicong School of Computer and Communication Engineering University of Science and Technology Beijing Beijing China Beijing Key Laboratory of Security and Privacy in Intelligent Transportation Beijing Jiaotong University Beijing China

ISBN: (纸本)9798400707308

Data-free generative modeling (GM) methods have gained much attention because pseudo data avoids security and privacy issues, and no amount limitation for storing real data in Class-Incremental Learning (CIL). However, GM-based methods have three drawbacks: un stable quality of generated image, semantic sifting, and domain bias for knowledge and distillation. To tackle with these problems, we propose an effective two-stage training method based on generative replay distillation, namely GA-RCT in this paper. It mainly consists of three modules. The Generative Adaptive Replay (GAR) module in stage one obtainsa well-trained generator for every task, then freezes generator for generating stable images, and trains feature extractor with mixture of real data and generated data. The Global Average Class (GAC) module is designed to reserve class prototypes to reduce semantic drift in the training stage. And the Generative Auxiliary Teaching (GAT) module in stage two performs generative distillation in dual-teacher to student way, for solving domain bias. Experiments on four datasets have shown that GA-RCT performs better than other generative replay methods. © 2024 Copyright held by the owner/author(s).

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Password-Based Authenticated Key Exchange for Cloud Storage from RLWE 24

Password-Based Authenticated Key Exchange for Cloud Storage ...

引用

2nd International Conference on Electronics, Computers and communication technology, CECCT 2024

作者： Xie, Gang Xu, Chunxiang Jiang, Changsong Han, Yu Dong, Xinfeng School of Computer Science and Engineering University of Electronic Science and Technology of China Sichuan Chengdu China University of Electronic Science and Technology of China Huzhou China School of Computer Science and Engineering and Technology of China University of Electronic Science and Technology of China Sichuan Chengdu China China and National Key Laboratory of Security Communication Sichuan Chengdu China

ISBN: (纸本)9798400710193

Cloud storage enables users to store, retrieve, and share large amounts of data in a flexible and convenient way. To guarantee only authorized users can access the data, authenticated key exchange is a feasible solution, which provides both user authentication and session key negotiation. In authenticated key exchange, passwords are the most widely used authenticatin factor due to their easy-operation. However, most password-authenticated key exchange rely on traditional cryptographic assumptions, such as the discrete logarithm problem or integer factorization, which are vulnerable to quantum attacks. In this paper, we propose a novel password-authenticated key exchange protocol dubbed PAKER based on Ring Learning With Errors problem(RLWE) for cloud storage, which is a Diffie-Hellman-like protocol and achieves post-quantum security. In PAKER, all computations are performed in a polynomial ring, where an approximate verification key is obtained through polynomial multiplication, followed by the application of a reconciliation function and an auxiliary modular function, a user and the server can negotiate a session key. We provide the formal security proof of PAKER. The comprehensive performance evaluation demonstrates that PAKER is efficient in accordance with computation and communication costs. © 2024 Copyright held by the owner/author(s).

关键词： Authentication Protocol

来源：评论

学校读者我要写书评

暂无评论

On l-th NMDS Codes∗

引用

Chinese Journal of Electronics 2025年第3期23卷 454-457页

作者： Hongxi Tong Wenhai Chen Fei Yu Department of Mathematics Shanghai University Shanghai China Science and Technology on Communication Security Laboratory Chengdu China

Linear codes with Singleton defects equivalent to 0, 1, 2 are studied, and they have good properties. They are called MDS codes, NMDS codes, NNMDS cdoes, respectively. In this paper, we study linear codes with large Singleton defects and these codes have same good properties with MDS codes, NMDS codes, NNMDS codes. We call them $\boldsymbol{l}$ -th NMDS codes. A series of iff conditions of $\boldsymbol{l}$ -th NMDS codes is presented. And we give an upper bound on length of $\boldsymbol{l}$ -th NMDS codes. In the last, some examples of $\boldsymbol{l}$ -th NMDS codes are given.

关键词： Codes Upper bound Linear codes Hamming weight

来源：评论

学校读者我要写书评

暂无评论

Research on Radar Anti-deceptive-jamming Method based on Characterization 16

Research on Radar Anti-deceptive-jamming Method based on Cha...

引用

16th International Conference on Signal Processing Systems, ICSPS 2024

作者： Lei, Lei Jiang, Meiqiu Li, JiaChen Guo, Shisheng Cui, Guolong Xia, Bijun Zhu, Lei School of Information and Communication Engineering University of Electronic Science and Technology of China Chengdu China National Key Laboratory of Electromagnetic Space Security Chengdu China

ISBN: (纸本)9781510689251

Deceptive jamming plays an increasingly important role in radar electronic countermeasures, and there is an urgent need to study new radar anti-deceptive jamming methods. In this paper, four characteristic factors, namely, amplitude variance, amplitude ratio, power ratio and number of correlated signals, are extracted based on the intra pulse characteristic differences between the target and deceptive jamming signals. Based on these four factors, a multi-domain cross-coupled neural network is proposed to distinguish between real targets and deceptive jamming. Finally, the simulation results show that this method has a significant effect on anti-deceptive jamming. © 2025 SPIE.

关键词： Radar countermeasures

来源：评论

学校读者我要写书评

暂无评论

Multi-scale network for single image deblurring based on ensemble learning module

引用

Multimedia Tools and Applications 2025年第11期84卷 9045-9064页

作者： Wu, WenBo Pan, Yun Su, Na Wang, JingTao Wu, ShaoChuan Xu, ZeKun Yu, YouJian Liu, YaPeng State Key Laboratory of Media Convergence and Communication Communication University of China Beijing100024 China School of Computer and Cyberspace Security Communication University of China Beijing100024 China School of Electronics and Information Engineering Harbin Institute of Technology Harbin150080 China College of Computer and Control Engineering Northeast Forestry University Harbin150040 China School of Computer and Information Engineering Tianjin Chengjian University Tianjin300384 China School of Computer Science and Technology Anhui University Hefei230601 China

In this paper, we have identified two primary issues with current multi-scale image deblurring methods. On the one hand, the blurring scale is ignored. On the other hand, the context information of images is not fully utilized, and the spatial detail information of images is lost. To better solve the above problems, on top of that, this paper proposes a multi-scale network for single image deblurring. The designed network contains two parts: 1) Ensemble Learning Model: The Weak Learning Model in this module connects multiple weak learners in series to increase network depth. Moreover, through multi-scale feature ensemble module, the output features and intermediate features of different weak learner are integrated, and finally strong learner is used to enhance the features with semantic information at different scales;2) Spatial Feature Enhancement Module: This module can preserve spatial detail information of images and compensate for the disadvantages of losing image spatial details due to continuous downsampling in the Ensemble Learning Model structure. Our proposed deblurring model has been extensively evaluated on several benchmark datasets, and the model achieves superior performance compared to the most advanced deblurring methods. In particular, the peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM) values are 33.57 dB and 0.965 on the GoPro dataset, respectively, with PSNR and SSIM improving by 0.49 dB and 0.002 over NAFNet, and 0.65 dB and 0.004 over Restormer. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Learning systems

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：