检索结果-内蒙古大学图书馆

Protecting machine learning systems using blockchain: solutions, challenges and future prospects

Multimedia Tools and applications 2024年 1-28页

作者： Hajlaoui, Rejab Dhahri, Salah Mahfoudhi, Sami Moulahi, Tarek Alotibi, Gaseb Department of Computer and Informtion Sciences College of Computer Science and Engineering University of Hail Hail City Saudi Arabia Laboratory of Advanced Multifunctional Materials and Technological Applications Faculty of Science and Technology of Sidi Bouzid University of Kairouan Sidi Bouzid Tunisia Department of Information and Technology Management Mediterranean School of Business South Mediterranean University Tunis Tunisia Department of Information Technology College of Computer Qassim University Buraidah Saudi Arabia Department of Computers and Information Technology Alwajah College University of Tabuk Tabuk Saudi Arabia

Machine learning-based systems have emerged as the primary means for achieving the highest levels of productivity and efficiency. They have become the most influential competitive factor for many technologies and business companies such as Cloud AI Companies (Google, Facebook…) and Health Care AI Companies (Tempus, Nanox …). However, privacy and security issues have become the biggest challenge facing all ML applications. These threats are known as Adversarial Machine Learning (AML). Their main goal is to maliciously manipulate training data and model parameters to obtain misleading results. To overcome these challenges, several research studies have proven that Blockchain, which relies on cryptographic technologies, constitutes a promising solution for securing ML applications. It provides high guarantees to make the system scalable, reliable and more secure. In this survey, we provide an overview of ML applications and the most hostile attacks they face. Then, we define the blockchain technique and its distinctive features. After that, we conduct a comprehensive and detailed survey of the latest and best blockchain-based research to address security issues in ML applications (integrity, confidentiality and availability). We analyze, compare and comment on all the proposed solutions. In the end, we discuss in detail the various challenges that hinder the adoption of blockchain technology and we extract the most important scientific trends that will serve as guidance and support for researchers in their future works. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Blockchain

来源：评论

学校读者我要写书评

暂无评论

GPU Side-Channel Attack Classification for Targeted Secure Shader Mitigation

引用

SN computer science 2024年第8期5卷 1148页

作者： Lungu, Nelson Patra, Sudhansu Shekhar Mishra, Manoj Ranjan Dash, Bibhuti Bhusan Sasmal, Gopal Charan Pattnayak, Parthasarathi Singh, Satyendr Gourisaria, Mahendra Kumar Department of Electrical and Electronics Engineering University of Zambia Lusaka Zambia School of Computer Applications Kalinga Institute of Industrial Technology (KIIT) Deemed to be University Bhubaneswar India MCA Department Trident Academy of Creative Technology Bhubaneswar India Computer Science and Engineering Department BML Munjal University Gurugram India School of Computer Engineering Kalinga Institute of Industrial Technology (KIIT) Deemed to be University Bhubaneswar India

Graphics processing units (GPUs) provide massively parallel processing capabilities, enabling accelerated computation across diverse applications. However, their parallel architecture and shared resources also introduce significant security vulnerabilities, especially side-channel attacks that can compromise sensitive data. This article presents a systematic classification of GPU side-channel attack vectors, including power analysis, timing analysis, electromagnetic analysis, and combined attacks. This taxonomy proposes countermeasures for the NVIDIA GeForce RTX 3090 GPU to mitigate these threats. The countermeasures encompass hardware techniques such as isolated execution environments, secure memory partitioning, and hardware random number generators to reduce the attack surface. Software methods, including constant-time algorithms, data-independent execution paths, masking, and oblivious memory access, are also implemented to eliminate timing and cache side channels. Comprehensive experiments demonstrate the effectiveness of the countermeasures, achieving up to 95% reduction in power analysis leakage, 90% in timing analysis, and 92% in electromagnetic analysis with acceptable performance overheads. The secure shader programming framework provides robust mitigation against GPU side-channel attacks while maintaining efficiency and compatibility with the RTX 3090 architecture. This work contributes vital insights and solutions for analysing GPU security vulnerabilities, designing focused countermeasures, and enabling confidential computing leveraging modern GPU capabilities. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2024.

关键词： Graphics processing units security Hardware countermeasures NVIDIA GeForce RTX 3090 Secure shader Side-channel attacks Software mitigation

来源：评论

学校读者我要写书评

暂无评论

Self-moving Window with Long-Short Term Memory based Anomaly Detection in Edge-Assisted IoT Environments

Self-moving Window with Long-Short Term Memory based Anomaly...

引用

2024 International Conference on Distributed Systems, computer Networks and Cybersecurity, ICDSCNC 2024

作者： Chanti, Yerrolla Adnan, Myasar Mundher Devi, R. Renuga Lakshmanan, M. Madhavi, Arelli Sr University School of Computer Science & Artificial Intelligence Warangal India The Islamic University College of Technical Engineering Department of Computers Techniques Engineering Najaf Iraq Srm Institute of Science and Technology Department of Computer Science and Applications Ramapuram India Nitte Meenakshi Institute of Technology Department of Eee Bengaluru India Gokaraju Rangaraju Institute of Engineering and Technology Department of Ai&ml Bachupally India

ISBN: (纸本)9798350375442

Internet of Things (IoT) systems is widely used in many applications which generate huge amounts of sensory signals that involves to signify the state of systems. However, due to changes in IoT environment impact the detection of anomalies accurately which lead to risk of losing information. To overcome this limitation, a Self-moving Window with Long Short-Term Memory (SW-LSTM) is proposed for accurate detection of anomalies. The self-moving window improved the ability of LSTM to identify subtle anomalies that evolve over time, leading to more accurate detection of unusual patterns or device behavior in IoT environments. Initially, the data about attacks and malicious nodes are acquired from the UNSW-15 dataset and preprocessed by score normalization technique. Then, the optimal set of features are selected by Particle Swarm Optimization (PSO) algorithm. Finally, the proposed SW-LSTM model detects the anomalies in edge assisted IoT network effectively. The experimental results of proposed method achieved accuracy of 93.4% which is greater when compared to existing detection approaches such as Recurrent Neural Network (RNN), Convolutional Neural Network (CNN). © 2024 IEEE.

关键词： Particle swarm optimization (PSO)

来源：评论

学校读者我要写书评

暂无评论

A novel three-phase hybrid cryptographic algorithm for data security

引用

International Journal of Information Technology (Singapore) 2024年 1-9页

作者： Pooja Shalu Shyla Sangwan, Yogita Shivahare, Basu Dev School of Computer Science Engineering and Technology Bennett University Greater Noida Uttar Pradesh Noida India Manav Rachna University Haryana Faridabad India Bharati Vidyapeeth’s College of Engineering Delhi India Department of Computer Science and Applications Kurukshetra University Haryana Kurukshetra India School of Computing Science and Engineering Galgotias University Greater Noida Uttar Pradesh Noida India

Cryptography is crucial for ensuring the data security in wireless sensor networks. The real data is altered or concealed using several hashing techniques before transmission across the network. But, these traditional techniques are not reliable in attaining security at multiple phases in the network. Thus, the given paper proposes a novel Three-Phase Hybrid Cryptographic (TPHC) algorithm which makes use of both symmetric and asymmetric cryptographic methods. It provides authentication and confidentiality using modified Rivest–Shamir–Adleman (RSA), advanced encryption standard (AES), and data encryption standard (DES). To provide integrity, TPHC is additionally paired with a revolutionary hashing method. The proposed hash function adheres to all four of the basic hashing properties: one-way, resistance to second pre-images, collision resistance, and avalanche effect. The results of this study demonstrate that TPHC has a good execution speed and a unique hashing technique that generates reduced message digests as compared with existing hashing algorithms. The TPHC algorithm excels in terms of computation speed, cipher text compactness, digest size, and energy efficiency within WSN’s. © Bharati Vidyapeeth's Institute of computer applications and Management 2024.

关键词： AES DES Hybrid cryptographic algorithm and hashing Modified-RSA

来源：评论

学校读者我要写书评

暂无评论

A Comprehensive Analysis of Cloud Computing in Wireless Communications 2

A Comprehensive Analysis of Cloud Computing in Wireless Comm...

引用

2nd International Conference on Disruptive Technologies, ICDT 2024

作者： Bansal, Anurag Sharma, Surbhi Shukla, Ajay Narayan Badhoutiya, Arti Research and Technology 13D Research & Strategy India Abes Engineering College Department of Computer Applications UP Ghaziabad India Computer Science & Engineering Graphic Era Hill University Uttarakhand Dehradun India Gla University Department of Electrical Engineering UP Mathura India

ISBN: (纸本)9798350371055

In the present-day digital landscape, cloud computing has grown to be a ubiquitous generation that offers numerous advantages to numerous businesses. Its software in wi-fi communications has had a mainly revolutionary effect, ultimately converting how data is transferred, considered, and saved. This looks at a thorough examination of how cloud computing has affected wireless communications, including its records, most important ideas, and programs. The primary phase of the object appears to be how cloud computing has changed over the years, moving from conventional on-premise computing to its current country, which includes a spread of service offerings and deployment methodologies. Subsequently, it explores the fundamental thoughts of cloud computing, which have virtualization, scalability, and on-demand provisioning, and talks about how these apply to wireless communications. The ability of cloud computing to grow network availability and dependability is one of its primary benefits for wi-fi communications. Using cloud-based computing and garage sources, network operators may additionally lower downtime and increase uptime for their services. They can also better manage unforeseen surges in community visitors and provide a higher user revel. The look additionally addresses how cloud computing helps the short rollout of the latest services and packages within the wireless communications industry. © 2024 IEEE.

关键词： Cloud Computing Data Storage Mobile Technology Network Security Wireless Communications

来源：评论

学校读者我要写书评

暂无评论

A comparative analysis of using Deep Learning and Machine Learning technologies for intrusion detection for effective network traffic analysis

A comparative analysis of using Deep Learning and Machine Le...

引用

International Conference on Circuit, Power and Computing Technologies (ICCPCT)

作者： Divya S B S. Prince Mary Department of Computer Science and Engineering Sathyabhama Institute of Science & Technology Chennai Department Of Computer Applications Mangalam College Of Engineering

ISBN: (数字)9798350372816

ISBN: (纸本)9798350372823

The broad utilization of interconnectivity and interoperability of processing frameworks have turned into an irreplaceable need to improve our everyday activities. All the while, it opens a way to exploitable weaknesses that work out positively past human control capacity. The weaknesses consider network protection instruments fundamental to accepting correspondence trade. Secure correspondence requires safety efforts to battle the dangers and necessities headways to safety efforts that counter advancing security dangers. So, emerging technologies like Deep Learning/ Machine Learning (DL/ML) come in handy for integration and better prediction. So, in this paper, we depict various deep learning/ML strategies used in detecting the imbalance in network traffic. The document also includes research that has been proposed on the topic of intrusion detection in network traffic over the previous four years (2020–2024) by a variety of research specialists. The main achievements of the security measures are recognized in this review together with their quantitative and qualitative success indicators. Additionally, as a guide for upcoming discoveries, this research study examines the major discoveries and the rationale for the lessons discovered.

关键词： Deep learning Performance evaluation Reviews Computational modeling Intrusion detection Telecommunication traffic Security

来源：评论

学校读者我要写书评

暂无评论

Cloud Top Management Role in Reducing Mobile Broadband Transmission Hazards and Offering Safety 3

Cloud Top Management Role in Reducing Mobile Broadband Trans...

引用

3rd International Conference on Advance Computing and Innovative Technologies in engineering, ICACITE 2023

作者： Manjunath, C.R. Sharma, Sachin Sharma, Aditi Sunil Kumar, M. Kelkar, Sandeep Vishal Deshmukh, Sheetal Department of Computer Science and Engineering Bangalore India Dev Bhoomi Uttarakhand University Department of Computer Science and Engineering Uttarakhand Dehradun India Parul University PIT Department of Computer Science and Engineering Gujarat Vadodara India Mohan Babu University Erstwhile SreeVidyanikethan Engineering College Department of Computer Science and Engineering AP Tirupathi India Prin L.N. Welingkar Institute of Management Development and Research India Yashwantrao Mohite Institute of Management Department of Computer Applications Karad India

ISBN: (纸本)9798350399264

In both the academic and industrial worlds, cloud technology is becoming a widely used research area. Both providers of cloud services and consumers profit from cloud computation. The literature has mostly focused on the security issues related to the cloud. This thorough literature study's objective is to assess the state of knowledge on dangers, difficulties, and privacy for public cloud. This studies from different countries at the areas of study that some well rising institutions are focusing between on 2010. and 2020. We chose 81 papers after carefully examining several attempts to solve the stated research issues. The study's results revealed seven serious security threats to internet data centres. The results revealed that a few of the extensively covered subjects with in chosen works included data modification and leaking. Network discontinuities, storing, and information inside the realm of public cloud were also recognised potential threats. The findings of this Study also shown that CSPs and internet users alike still have difficulties when reassessing client data. The cryptocurrency was identified in our research report as an innovation activitythat would help to allay security worries. The SLR findings reveal a few things that should be done in efforts to provide categorization, legitimacy, and information availability in the present. © 2023 IEEE.

关键词： Industrial research

来源：评论

学校读者我要写书评

暂无评论

Performance Enhancement of a Vertical Axis Wind Turbine (NACA 25112) Using Deflector Plates as an Augmenter 2nd

Performance Enhancement of a Vertical Axis Wind Turbine (NAC...

引用

2nd International Conference on Advances in Mechanical engineering and Material science, ICAMEMS 2023

作者： Kumar, Karthik Selva Jaseel, Mohammed Brintha, R. Devi, Aruna Jaiswal, Krishna Kumar Sangmesh, B.M. Suresh, A. Balaji, P.S. Department of Green Energy Technology Pondicherry University Puducherry India Department of Computer Applications Saradha Gangadharan College Puducherry India Department of Computer Science and MIS OCMT Barka Oman Department of Mechanical Engineering NIT Warangal Hanamkonda India Department of Mechanical Engineering NIT Rourkela Rourkela India

ISBN: (纸本)9789819956128

In order to overcome the self-starting issues in vertical axis wind turbines (VAWT), it is proposed to introduce the deflector plates as a power augmenter. In this study, a computational analysis on a wind turbine rotor with three-bladed NACA 25112 has been performed to understand the significance of power augmenters and their impact on the overall performance of the wind energy conversion system. The rotor motion of the proposed model has been simulated by introducing a sliding mesh technique with reference to the SST-turbulence model. The simulation has been performed at different flow velocities ranging from 4 to 7 m/s to understand the effects of power augmenters with respect to the enhancement in the initial torque. The observed results showed a significant improvement in the enhancement of initial torque, the aerodynamic behavior of the airfoils, and the power coefficient of the wind turbine. © 2024, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Vertical axis wind turbine

来源：评论

学校读者我要写书评

暂无评论

GS-FFO Based Meta-Heuristic Clustering for Discovering Colossal Patterns in High-Dimensional Data 2nd

GS-FFO Based Meta-Heuristic Clustering for Discovering Colos...

引用

2nd Congress on Control, Robotics, and Mechatronics, CRM 2024

作者： Reddy, T. Sreenivasula Sathya, R. Nuka, Mallikharjuna Rao Research Scholar Department of Computer Science and Engineering Faculty of Engineering and Technology Annamalai University Annamalai Nagar Tamil Nadu Chidambaram608002 India Assistant Professor Department of Computer Science and Engineering Faculty of Engineering and Technology Annamalai University Annamalai Nagar Tamil Nadu Chidambaram608002 India Professor Department of Computer Applications Annamacharya Institute of Technology & Sciences YSR Kadapa Andhra Pradesh Rajampet516115 India

ISBN: (纸本)9789819768097

Colossal pattern mining, which involves deriving important relationships and insights from large and complex datasets, is an essential part of data analysis. Extracting association rules that satisfy preset minimum support and confidence criteria from a given database is the aim of association rule mining (ARM). Nevertheless, ARM often generates an unusually large number of association rules that are hard for end users to comprehend or validate, which lessens the significance of data mining discoveries. Based on the clustering methodology, this research study presents a new mining strategy. The process begins with a method called the Expressed Data Matrix (EDM), which transforms unstructured data into a structured format and enhances the quality and organisation of the information. Primary data cleaning processes are used to eliminate noise and inconsistencies from the data in order to further improve it. To address high-dimensionality issues and boost mining productivity, dimensionality reduction—which lowers the dataset to its most valuable features—is accomplished using Eigen-Principal Component Analysis (PCA). The core of this study is the hybridised clustering approach, or GS-FFO, which combines the Fruit Fly Optimisation technique (FOA) with Glow Swarm Optimisation (GSO). This technique improves the quality of clustering and reduces the probability of local optima convergence by integrating the benefits of both approaches. The suggested GS-FFO model outperforms with four datasets, including accident, pumbs*, retail, and yeast, with an accuracy of up to 98.5%. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Dimensionality reduction

来源：评论

学校读者我要写书评

暂无评论

Examining the Impact of Machine Learning on Streamlining Data Analysis in Real-Time 2

Examining the Impact of Machine Learning on Streamlining Dat...

引用

2nd International Conference on Disruptive Technologies, ICDT 2024

作者： Singh, Rashi Goel, Mukulit Bahuguna, Pankaj Kumar, Anuj Jain, Sachin Mvn University Department of Computer Science & Applications Haryana Palwal India Ajay Kumar Garg Engineering College Department of Mca UP Ghaziabad India Ajay Kumar Garg Engineering College Department of Cse UP Ghaziabad India

ISBN: (纸本)9798350371055

6G wireless conversation networks are anticipated to be one of the number one subsequent-era network answers to provide services and studies to users. So that you can fully understand the sizeable data throughput and extremely low latency benefits associated with 6G, tailored resource allocation algorithms are utilized to efficiently manage and allocate resources to some of the distinctive users of the network. This summary review presents useful resource allocation algorithms for 6G networks, consisting of OFDMA, TDMA, SDLMA, TSMA, and MUSA. Reading the pros and cons of every set of rules well-known shows the most effective answers for the numerous demands of 6G users. It's also mentioned how container orchestration strategies along with Cabernets can be used to decorate useful resource orchestration in 6G networks. With the excessive demand for efficient aid allocation algorithms, this review will provide a critical framework for the development of 6G conversation networks. © 2024 IEEE.

关键词： Automation Data Analysis Machine Learning Real-Time Streamlining

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：