检索结果-内蒙古大学图书馆

RobustPFL: Robust Personalized Federated Learning

IEEE Transactions on Dependable and Secure computing 2025年

作者： Chen, Guorong Wang, Wei Wu, Yufang Li, Chao Xu, Guangquan Ji, Shouling Li, Tao Shen, Meng Han, Yufei Xi'an Jiaotong University Ministry of Education Key Lab for Intelligent Networks and Network Security Xi'an China Beijing Jiaotong University 3 Shangyuancun Beijing Key Laboratory of Security and Privacy in Intelligent Transportation Beijing100044 China Tianjin University Tianjin Key Laboratory of Advanced Networking College of Intelligence and Computing Tianjin300350 China Zhejiang University College of Computer Science and Technology Hangzhou310027 China ITAI Haihe Lab Tianjin China Nankai University College of Computer Science Tianjin300350 China Beijing Institute of Technology School of Cyberspace Security Beijing100081 China Peng Cheng Laboratory Cyberspace Security Research Center Shenzhen518066 China INRIA Bretagne Rennes35042 France

Conventional federated learning (FL) coordinated by a central server focuses on training a global model and protecting the privacy of clients' training data by storing it locally. However, the statistical heterogeneity hinders the global model from adapting to the non-IID distributions among clients. Moreover, untrusted and unreliable central servers and malicious clients may compromise model integrity and availability, thus degrading the robustness of FL. To address these challenges, we present RobustPFL, a decentralized personalized federated learning (PFL) approach that combines α-based Layer-position Normalized Similarity (α-LNS) and local collaborative training to improve personalized performance while utilizing a blockchain-based committee mechanism to coordinate the aggregation process, thereby achieving high personalized accuracy and robustness. Extensive experiments show that our RobustPFL approach outperforms multiple algorithms, including Local training, FedAvg, FedReptile, Per-FedAvg, FedBN, and SPFL, on MNIST, CIFAR10, EMNIST, and N-BaIoT datasets in four non-IID settings. We also evaluate RobustPFL's effectiveness against attacks - poisoning attacks and free-riding attacks. Particularly, for three prevalent poisoning attacks (backdoor, label flipping, and model poisoning attacks), we compare non-defensive (FedAvg) and defensive (Krum, trimmed mean, Bulyan, FedBN, FLAME, and FangTrmean) methods with our proposed RobustPFL. The results show that our approach achieves significant defensive effects. © 2004-2012 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

HiCoCS: High Concurrency Cross-Sharding on Permissioned Blockchains

arXiv

引用

arXiv 2025年

作者： Yang, Lingxiao Dong, Xuewen Wan, Zhiguo Lu, Di Zhang, Yushu Shen, Yulong School of Computer Science and Technology Xidian University The Engineering Research Center of Blockchain Technology Application and Evaluation Ministry of Education China Shaanxi Key Laboratory of Blockchain and Secure Computing Xi’an710071 China Zhejiang Lab Zhejiang Hangzhou311121 China School of Computer Science and Technology Xidian University China Shaanxi Key Laboratory of Network and System Security Xi’an710071 China College of Computer Science and Technology Nanjing University of Aeronautics and Astronautics Jiangsu Nanjing210016 China

As the foundation of the Web3 trust system, blockchain technology faces increasing demands for scalability. Sharding emerges as a promising solution, but it struggles to handle highly concurrent cross-shard transactions (CSTxs), primarily due to simultaneous ledger operations on the same account. Hyperledger Fabric, a permissioned blockchain, employs multi-version concurrency control for parallel processing. Existing solutions use channels and intermediaries to achieve cross-sharding in Hyperledger Fabric. However, the conflict problem caused by highly concurrent CSTxs has not been adequately resolved. To fill this gap, we propose HiCoCS, a high concurrency cross-shard scheme for permissioned blockchains. HiCoCS creates a unique virtual sub-broker for each CSTx by introducing a composite key structure, enabling conflict-free concurrent transaction processing while reducing resource overhead. The challenge lies in managing large numbers of composite keys and mitigating intermediary privacy risks. HiCoCS utilizes virtual sub-brokers to receive and process CSTxs concurrently while maintaining a transaction pool. Batch processing is employed to merge multiple CSTxs in the pool, improving efficiency. We explore composite key reuse to reduce the number of virtual sub-brokers and lower system overhead. Privacy preservation is enhanced using homomorphic encryption. Evaluations show that HiCoCS improves cross-shard transaction throughput by 3.5-20.2 times compared to the baselines. Copyright © 2025, The Authors. All rights reserved.

关键词： Concurrency control

来源：评论

学校读者我要写书评

暂无评论

HiCoCS: High Concurrency Cross-Sharding on Permissioned Blockchains

引用

IEEE Transactions on Computers 2025年

作者： Yang, Lingxiao Dong, Xuewen Wan, Zhiguo Lu, Di Zhang, Yushu Shen, Yulong Xidian University School of Computer Science and Technology China Ministry of Education Engineering Research Center of Blockchain Technology Application and Evaluation China Shaanxi Key Laboratory of Blockchain and Secure Computing Xi’an710071 China Zhejiang Lab Zhejiang Hangzhou311121 China Shaanxi Key Laboratory of Network and System Security Xi’an710071 China Jiangxi University of Finance and Economics School of Computing Nanchang330013 China Jiangxi University of Finance and Economics Nanchang330013 China

关键词： Scalability

来源：评论

学校读者我要写书评

暂无评论

A strategy for avoiding energy holes based on data fusion in wireless sensor network

A strategy for avoiding energy holes based on data fusion in...

引用

2012 International Conference on Electrical Insulating Materials and Electrical Engineering, EIMEE 2012

作者： Song, Xiaoying Wen, Tao Guo, Quan Wu, Di China Liaoning Key Lab of Neusoft Security and Computing Technology Dalian Neusoft Information Institute Dalian Liaoning China

ISBN: (纸本)9783037854518

Energy holes problem is one of key issues for wireless sensor networks (WSN). At present, the resolved strategies mainly focus on non-uniform node distribution and adjusting transmission power. However, some costs are incurred by non-uniform node distribution. With the number of nodes increasing, the total number of nodes grows exponentially, and there is a need to improve physical conditions when we increase transmission power significantly in large class of network model;To overcome the limitation, we propose the strategy of data fusion, whose reason is that it is not limited to the size of network, and does not need complicated physical conditions. In this paper, we first establish a network model Multiple-Layer and Multiple-Clusters in corona model (MLMC) to analyze the loading energy consumption. What's more, we consider energy consumption when network is constructed. Furthermore, we fully introduce our strategy of data fusion, and prove two theorems to balance energy consumption for resolving energy holes problem. Finally, we compare and analyze the total energy consumption and network lifetime between before and after data fusion. The experimental results demonstrate the performance advantages of our approach. The best is to read these instructions and follow the outline of this text. © (2012) Trans Tech Publications, Switzerland.

关键词： Energy utilization

来源：评论

学校读者我要写书评

暂无评论

The Algorithm of Short Message Hot Topic Detection Based on Feature Association

引用

Information technology Journal 2009年第2期8卷 236-240页

作者： Sun, Qindong Wang, Qian Qiao, Hongli School of Computer Science and Engineering Xi'an University of Technology Xi'an 710048 China Network Computing and Security Key Lab of Shaanxi Xi'an University of Technology Xi'an 710048 China

Aiming at the mobile short message (SMS) hot topic extraction, the text features and statistical regularity of SMS are analyzed in this study. The formal description of SMS hot topic is given and an algorithm of SMS hot topic extraction based on feature association analysis is proposed. According to the proposed algorithm, feature words of SMS can be clustered into different word bags by calculating the association degree of these feature words and the hot topic can be identified by means of word bags matching. Experiments results show that the proposed algorithm can detect the hot topic in the SMS messages effectively, which is useful to the analysis of SMS popular sentiments. © 2009 Asian network for Scientific Information.

关键词： Data mining

来源：评论

学校读者我要写书评

暂无评论

PosFuzz:augmenting greybox fuzzing with effective position distribution

引用

Cybersecurity 2023年第4期6卷 123-143页

作者： Yanyan Zou Wei Zou JiaCheng Zhao Nanyu Zhong Yu Zhang Ji Shi Wei Huo Institute of Information Engineering Chinese Academy of SciencesBeijingChina School of Cyber Security University of Chinese Academy of SciencesBeijingChina State Key Lab of Processors Institute of Computing TechnologyChinese Academy of SciencesBeijingChina Key Laboratory of Network Assessment Technology Chinese Academy of SciencesBeijingChina Beijing Key Laboratory of Network Security and Protection Technology BeijingChina Zhongguancun Laboratory BeijingChina

Mutation-based greybox fuzzing has been one of the most prevalent techniques for security vulnerability discovery and a great deal of research work has been proposed to improve both its efficiency and ***-based greybox fuzzing generates input cases by mutating the input seed,i.e.,applying a sequence of mutation operators to randomly selected mutation positions of the ***,existing fruitful research work focuses on scheduling mutation operators,leaving the schedule of mutation positions as an overlooked aspect of fuzzing *** paper proposes a novel greybox fuzzing method,PosFuzz,that statistically schedules mutation positions based on their historical *** makes use of a concept of effective position distribution to represent the semantics of the input and to guide the *** first utilizes Good-Turing frequency estimation to calculate an effective position distribution for each mutation *** then leverages two sampling methods in different mutating stages to select the positions from the *** have implemented PosFuzz on top of AFL,AFLFast and MOPT,called Pos-AFL,-AFLFast and-MOPT respectively,and evaluated them on the UNIFUZZ benchmark(20 widely used open source programs)and LAVA-M *** result shows that,under the same testing time budget,the Pos-AFL,-AFLFast and-MOPT outperform their counterparts in code coverage and vulnerability discovery *** with AFL,AFLFast,and MOPT,PosFuzz gets 21%more edge coverage and finds 133%more paths on *** also triggers 275%more unique bugs on average.

关键词： Greybox fuzzing Mutation position Mutation operator Code coverage Vulnerability discovery

来源：评论

学校读者我要写书评

暂无评论

A prior embedding-driven architecture for long distance blind iris recognition

引用

Biomedical Signal Processing and Control 2025年 109卷

作者： Qi Xiong Xinman Zhang Jun Shen International College Hunan University of Arts and Sciences Changde 415000 China School of Automation Science and Engineering Faculty of Electronic and Information Engineering MOE Key Lab for Intelligent Networks and Network Security Xi’an Jiaotong University Xi’an 710049 Shaanxi China School of Computing and Information Technology University of Wollongong Wollongong NSW 2522 Australia

Blind iris images, caused by unknown degradation during the process of iris recognition at long distances, often lead to decreased iris recognition rates. Currently, limited literature addresses this issue. To tackle this challenge, we propose a prior embedding-driven architecture for long-distance blind iris recognition. Our approach introduces a blind iris image restoration network named Iris-PPRGAN. To effectively restore the texture of blind iris images, Iris-PPRGAN incorporates a Generative Adversarial network (GAN) as a Prior Decoder and a Deep Neural network (DNN) as the encoder. To enhance iris feature extraction, we also developed a robust iris classifier by modifying the bottleneck module of InsightFace, referred to as Insight-Iris. Initially, a low-quality blind iris image is restored using Iris-PPRGAN, and the restored image is subsequently recognized through Insight-Iris. Experimental results on the public CASIA-Iris-distance dataset demonstrate that our method outperforms state-of-the-art blind iris restoration techniques. Specifically, the recognition rate for long-distance blind iris images improves from 80.77 % (without restoration) to 90.38 % after applying our method, reflecting an approximate ten-percentage-point improvement. These results, validated both quantitatively and qualitatively, underscore the effectiveness of our approach in addressing the challenges of long-distance blind iris recognition.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Design of distribution transformer monitoring system based on LPC2103

引用

Journal of Convergence Information technology 2012年第21期7卷 120-126页

作者： Sun, Qindong Wang, Qian Wang, Nan Network Computing and Security Key Lab of Shaanxi Xi'an University of Technology Xi'an 710048 China

Aimed at satisfying the monitoring requirement of distribution transformer, a kind of distribution transformer monitoring system based on LPC2103 is designed. Hardware module and data processing program of the monitoring system are introduced. The SD card of large capacity is used to store data collected by the supervising system, which could support the detail analysis of running status of distribution transformer. A simplified design of electricity cut examination is adopted to decrease the complexity and cost of the system. The system has been used in truth circumstance and the performance of system is satisfactory.

关键词： Monitoring

来源：评论

学校读者我要写书评

暂无评论

An algorithm for automatic photo typesetting based on kinship

引用

Journal of Information and Computational Science 2014年第13期11卷 4727-4736页

作者： Sun, Qindong Wang, Yan Tian, Minying Wang, Qian Zhang, Jingpeng Key Lab for Network Computing and Security of Shaanxi Province Xi'an University of Technology Xi'an China

Photo typesetting belongs to the rectangular packing problems, currently the algorithms of rectangular packing problems can be divided into two categories: heuristic methods and modern intelligent based methods. For both of these methods, the similarity of contents do not be well considered, the packing order of the rectangular pieces are chaos. Therefore, this drawback will cause problems for photo typesetting. In this paper we propose a hybrid algorithm for the photo typesetting problem based on kinship. According to our algorithm, the kinship of different photos is calculated by their absolute paths, and then all photos are packed based on the distance of their kinship. Compared with the existing methods, according to the computation of different scales of tests, our algorithm shows faster typesetting speed and higher utilization ratios. This algorithm could solve the sort problem caused by chaos typesetting photos of different customers after lots of photos were typeset. Our algorithm could greatly improve the practical application of photo typesetting. Copyright © 2014 Binary Information Press.

关键词： Heuristic methods

来源：评论

学校读者我要写书评

暂无评论

Modeling and analysis of the proactive worm in unstructured Peer-to-Peer network

引用

Journal of Convergence Information technology 2010年第5期5卷 111-117页

作者： Sun, Qindong Wang, Qian Ren, Jie Network Computing and Security Key Lab of Shaanxi Xi'an University of Technology Xi'an 710048 China

With the popularity of P2P (Peer-to-Peer) network applications, active P2P worms become one of the most important newly threat of the network security. In this paper, the propagation process of active P2P worms are studied by a dynamic model over the overlay topology which modeled as a power-law undirected graph. Firstly, GLP power law generator is used to generate power law topologies to represent unstructured P2P networks. Then the propagation of P2P worms is modeled over the topology. Specifically, discrete-time is adopted to conduct recursion and approximation to describe propagation of proactive P2P worm. In our model, all the probabilities which would happen to all kinds of peers are considered in full. Therefore, both the whole network's status and the single node's action are taken into consideration. So the model we constructed is more likely the truth. Finally, the simulation studies are carried out to prove that the simulation results match the mathematical model well.

关键词： Peer to peer networks

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：