检索结果-内蒙古大学图书馆

28th IEEE Symposium on computers and Communications, ISCC 2023

作者： Wang, Zhiye Liu, Baisong Lin, Chennan Zhang, Xueyuan Hu, Ce Qin, Jiangcheng Luo, Linze Ningbo University Faculty of Electrical Engineering and Computer Science Ningbo China

ISBN: (纸本)9798350300482

Deep learning based recommender systems(DLRS) as one of the up-And-coming recommender systems, and their robustness is crucial for building trustworthy recommender systems. However, recent studies have demonstrated that DLRS are vulnerable to data poisoning attacks. Specifically, an unpopular item can be promoted to regular users by injecting well-crafted fake user profiles into the victim recommender systems. In this paper, we revisit the data poisoning attacks on DLRS and find that state-of-The-Art attacks suffer from two issues: user-Agnostic and fake-user-unitary or target-item-Agnostic, reducing the effectiveness of promotion attacks. To gap these two limitations, we proposed our improved method Generate Targeted Attacks(GTA), to implement targeted attacks on vulnerable users defined by user intent and sensitivity. We initialize the fake users by adding seed items to address the cold start problems of fake users so that we can implement targeted attacks. Our extensive experiments on two real-world datasets demonstrate the effectiveness of GTA. © 2023 IEEE.

关键词： Recommender systems

来源：评论

学校读者我要写书评

暂无评论

A 9.5ms-Latency 6.2μJ/Inference Spiking CNN for Patient-Specific Seizure Detection

A 9.5ms-Latency 6.2μJ/Inference Spiking CNN for Patient-Spe...

引用

2023 IEEE Biomedical Circuits and Systems Conference, BioCAS 2023

作者： Muneeb, Abdul Mehrotra, Shreshth Kassiri, Hossein York University Department of Electrical Engineering and Computer Science Toronto Canada

ISBN: (纸本)9798350300260

We present the design, optimization, hardware implementation (generic and neuromorphic), and performance validation of regular and spiking convolutional neural networks (CNN and SCNN) for patient-specific seizure detection. The CNN's design is optimized using a neural architecture search, which resulted in a 7-layer network with one-dimensional (both horizontal and vertical) kernels that prove to be highly effective in capturing input's spatio-temporal features. The SCNN is developed using shadow training, and the conversion process is optimized for maximum model fidelity, which results in a © 2023 IEEE.

关键词： computational cost electroencephalogram (EEG) energy-efficient epilepsy neuromorphic seizure detection spiking neural network (SNN)

来源：评论

学校读者我要写书评

暂无评论

Secondary User Power Detection Based on UTAMP 18

Secondary User Power Detection Based on UTAMP

引用

18th IEEE Conference on Industrial Electronics and Applications, ICIEA 2023

作者： Qiu, Keyan Jin, Ming Yuan, Tianxi Xu, He Zha, Ziwei Ningbo University Faculty of Electrical Engineering and Computer Science Ningbo China

ISBN: (纸本)9798350312201

Integrated sensing and communication (ISAC) refers to a new information processing technology that achieves collaborative sensing and communication functions based on resource or information sharing between software and hardware. Wireless sensing technology is an important foundation for communication-aware sensing integration. This paper utilizes passive sensing technology, where the sensing nodes perceive and detect secondary users (SUs). The sensing nodes use the unitary transformation approximate message passing (UTAMP) algorithm to recover the received signal power sent by SUs, enabling detection of SUs. Simulation results demonstrate that the proposed algorithm has good detection performance, and complexity analysis of this method is conducted. But as no other papers have previously researched similar methods, there are no comparison methods. © 2023 IEEE.

关键词： Message passing

来源：评论

学校读者我要写书评

暂无评论

Behaviors Speak More: Achieving User Authentication Leveraging Facial Activities via mmWave Sensing 24

Behaviors Speak More: Achieving User Authentication Leveragi...

引用

22nd ACM Conference on Embedded Networked Sensor Systems, SenSys 2024

作者： Jiang, Chenxu Yu, Sihan Fu, Jingjing Lin, Chun-Chih Zhu, Huadi Ma, Xiaolong Li, Ming Guo, Linke Department of Electrical and Computer Engineering Clemson University ClemsonSC United States Department of Computer Science and Engineering The University of Texas at Arlington ArlingtonTX United States

ISBN: (纸本)9798400706974

Human faces have been widely adopted in many applications and systems requiring a high-security standard. Although face authentication is deemed to be mature nowadays, many existing works have demonstrated not only the privacy leakage of facial information but also the success of spoofing attacks on face biometrics. The critical reason behind this is the failure of liveness detection in biometrics. This work advances most biometric-based user authentication schemes by exploring dynamic biometrics (human facial activities) rather than traditional static biometrics (human faces). Inspired by observations from psychology, we propose the mmFaceID to leverage humans' dynamic facial activities when performing word reading for achieving robust, highly accurate, and effective user authentication via mmWave sensing. By addressing a series of technical challenges of capturing micro-level facial muscle movements using a mmWave sensor, we build a neural network to reconstruct facial activities via estimated expression parameters. Then, unique features can be extracted to enable robust user authentication regardless of relative distances and orientations. We conduct comprehensive experiments on 23 participants to evaluate mmFaceID in terms of distances/orientations, length of word lists, occlusion, and language backgrounds, demonstrating an authentication accuracy of 94.7%. We also extend our evaluation in a real IoT scenario. By speaking real IoT commends, the average authentication accuracy can reach up to 92.28%. © 2024 Copyright is held by the owner/author(s).

关键词： Biometrics

来源：评论

学校读者我要写书评

暂无评论

Lag Synchronization and Lag H∞ Synchronization for Multiweighted Coupled Reaction-Diffusion Neural Networks Suffering Topology Attacks

引用

IEEE Transactions on Control of Network Systems 2025年第2期12卷 1488-1499页

作者： Wang, Jin-Liang Zhao, Xin-Yu Ren, Shun-Yan Huang, Tingwen Tiangong University Tianjin Key Laboratory of Autonomous Intelligence Technology and Systems School of Computer Science and Technology Tianjin300387 China Tiangong University School of Electronics & Information Engineering Tianjin300387 China University of Electronic Science and Technology of China School of Mechanical and Electrical Engineering Chengdu611731 China Shenzhen University of Advanced Technology Faculty of Computer Science and Control Engineering Shenzhen518055 China

This paper mainly discusses two kinds of coupled reaction-diffusion neural networks (CRNN) under topology attacks, that is, the cases with multistate couplings and with multiple spatial-diffusion couplings. On one hand, by adopting state feedback control scheme, the problem of the lag synchronization for the multistate coupled CRNN is addressed. Moreover, the influences of external disturbances for the lag synchronization of the CRNN with multistate couplings are also taken into consideration. On the other hand, several lag and lag H∞ synchronization conditions for CRNN with multiple spatial-diffusion couplings are derived under the help of the Kronecker product and the state feedback control method. Finally, two numerical examples are provided to verify the correctness of the derived criteria. © 2014 IEEE.

关键词： State feedback

来源：评论

学校读者我要写书评

暂无评论

Extended Pulsation Attack Against Kalman Filter Driven BLDC Motor Control System 13

Extended Pulsation Attack Against Kalman Filter Driven BLDC ...

引用

13th International Conference on Advanced computer Information Technologies, ACIT 2023

作者： Boiko, Yuri Kiringa, Iluju Yeap, Tet University of Ottawa Electrical Engineering and Computer Science Department OttawaON Canada

The extended version of the induced pulsation attackis introduced through modeling and simulation, targeting a closed-loop-controlled architecture of a Brushless DC (BLDC) motor driven by a Kalman filter. Previously, the induced pulsations in the attack were created by a constant gradient linear function of the distortion coefficient employed by the attacker. In the updated version, the attack persists even when interruptions are introduced to the linear growth of the distortion function, followed by resetting it. The objective of the presented study isto assess the potential for enhancing the attack's effectiveness through this modification. The resulted findings confirm thattransforming the constant gradient linear function into a saw-tooth pattern leads to periodic repetitions of induced pulsations. Consequently, the attack can be extended indefinitely. Moreover,the attacker gains greater control over attack outcome by utilizing additional parameters introduced through the saw-like functionality, such as frequency and amplitude of saw-function. The study demonstrates two advantageous synchroniza-tion modes with the saw-function for the attacker: (1) modulatingcurrent, voltage, and angular speed in phase with the saw-function, providing a stronger attack but reduced covertness;(2) synchronizing current and voltage patterns with the first sub-harmonic of the saw-function while maintaining fundamental harmonic pace for angular speed modulation, resulting in improved covertness and consistent mechanical strains. Extended pulsation attack signatures collected in current, voltage, and angular speed domains aid in early detection through pattern recognition methods. © 2023 IEEE.

关键词： Controllers

来源：评论

学校读者我要写书评

暂无评论

High-Performance Flow Classification of Big Data Using Hybrid CPU-GPU Clusters of Cloud Environments

引用

Tsinghua science and Technology 2024年第4期29卷 1118-1137页

作者： Azam Fazel-Najafabadi Mahdi Abbasi Hani H.Attar Ayman Amer Amir Taherkordi Azad Shokrollahi Mohammad R.Khosravi Ahmed A.Solyman Department of Computer Engineering Faculty of EngineeringBu-Ali Sina UniversityHamedan ***Iran Department of Energy Engineering Zarqa UniversityZarqa 13132Jordan Department of Informatics University of OsloOslo 0316Norway Department of Computer Science MalmöUniversityMalmö20506Sweden Shandong Provincial University Laboratory for Protected Horticulture Weifang University of Science and TechnologyWeifang 261100China Department of Electrical and Electronics Engineering NişantaşıUniversityIstanbul 34481742Türkiye

The network switches in the data plane of Software Defined Networking (SDN) are empowered by an elementary process, in which enormous number of packets which resemble big volumes of data are classified into specific flows by matching them against a set of dynamic rules. This basic process accelerates the processing of data, so that instead of processing singular packets repeatedly, corresponding actions are performed on corresponding flows of packets. In this paper, first, we address limitations on a typical packet classification algorithm like Tuple Space Search (TSS). Then, we present a set of different scenarios to parallelize it on different parallel processing platforms, including Graphics Processing Units (GPUs), clusters of Central Processing Units (CPUs), and hybrid clusters. Experimental results show that the hybrid cluster provides the best platform for parallelizing packet classification algorithms, which promises the average throughput rate of 4.2 Million packets per second (Mpps). That is, the hybrid cluster produced by the integration of Compute Unified Device Architecture (CUDA), Message Passing Interface (MPI), and OpenMP programming model could classify 0.24 million packets per second more than the GPU cluster scheme. Such a packet classifier satisfies the required processing speed in the programmable network systems that would be used to communicate big medical data.

关键词： OpenMP Compute Unified Device Architecture(CUDA) Message Passing Interface(MPI) packet classification medical data tuple space algorithm Graphics Processing Unit(GPU)cluster

来源：评论

学校读者我要写书评

暂无评论

SAR UAV for soil moisture estimation 8

SAR UAV for soil moisture estimation

引用

8th Asia-Pacific Conference on Synthetic Aperture Radar, APSAR 2023

作者： Gleich, Dušan University of Maribor Faculty of Electrical Engineering and Computer Science Maribor Slovenia

ISBN: (纸本)9798350393590

Unmanned Aerial Vehicles (UAVs) equipped with Synthetic Aperture Radar (SAR) have revolutionized soil moisture estimation. UAV-based SAR offers high-resolution imaging, allowing detailed analysis at the field level, and greater temporal flexibility compared to satellite-based systems. The active microwave sensing capability of UAV-based SAR allows for all-weather and day-night operation, providing consistent and reliable soil moisture estimations. This paper presents a simple, lightweight, low consumption Frequency Modulated Continuous Wave (FMCW) radar attached to a hexacopter. The radar is designed to generate and acquire data in HH and VV polarization. A simple experiment is presented to estimate a soil moisture using the Dubois and the Integral Equation Model (IEM) for soil moisture detection. A custom designed convolutional network with attention mechanism for soil moisture estimation is presented. ©2023 IEEE.

关键词： Soil moisture

来源：评论

学校读者我要写书评

暂无评论

A Constrained Multi-Objective Evolutionary Algorithm Based on an Infeasible Archive

A Constrained Multi-Objective Evolutionary Algorithm Based o...

引用

2023 China Automation Congress, CAC 2023

作者： Ma, Yongjian Shi, Xuhua Ningbo University Faculty of Electrical Engineering and Computer Science Ningbo China

ISBN: (纸本)9798350303759

In the field of engineering design, there is a class of constrained multi-objective optimization problems where the optimal solutions are often found at the constraint boundaries. However, effectively utilizing the infeasible solutions at these boundaries poses a challenge in solving these problems. To address this issue, this paper proposes a constrained multi-objective evolutionary algorithm based on an infeasible archive (CIA). CIA utilizes two populations: one considers both objectives and constraints, while the other ignores all constraints. Additionally, an archive is implemented to store non-dominated infeasible solutions. To improve the search coverage in the initial stage and accelerate convergence in the later stage, a strategy is implemented that reduces infeasible solutions using the infeasible archive. Furthermore, a neighborhood-based offspring generation mechanism is developed, which combines feasible and infeasible solutions to enhance the ability of the algorithm to explore constraint boundaries. Experimental results from three test suites demonstrate that the proposed algorithm performs competitively when compared to four representative algorithms. © 2023 IEEE.

关键词： Evolutionary algorithms

来源：评论

学校读者我要写书评

暂无评论

Classifying wireless IOT ICU traffic with machine learning models 19

Classifying wireless IOT ICU traffic with machine learning m...

引用

19th International Conference on Future Networks and Communications, FNC 2024 / 21st International Conference on Mobile Systems and Pervasive Computing, MobiSPC 2024 / 14th International Conference on Sustainable Energy Information Technology, SEIT 2024

作者： Sibai, Fadi N. Sibai, Ahmad Electrical & Computer Engineering Dept. Gulf University for Science and Technology Mishref Kuwait Computer Science & Engineering Dept. University of South Florida TampaFL33620 United States

In this work, we classified the wireless internet of things (IoT) traffic of the IoT Health intensive care unit (IHI) dataset which belongs to three general classes: patient monitoring, environment monitoring, and network attack. We trained and tested 7 machine learning (ML) models with Orange 3 including kNN, Decision Tree (tree), SVM, Random Forest (RF), Neural Network (NN), Gradient Boosting (GB), and AdaBoost (AB). With the original dataset, 5 ML models performed perfect classification. After pruning the dataset columns by keeping the features with the highest correlations with the label in the dataset, good classifications were obtained with only 4 TCP/IP features by the Gradient Boosting, kNN, and RF models with MSEs in the range 0.008-0.011, and R2s in the range 0.978-0.984. With only 6 MQTT features, Gradient Boosting, RF, Tree, and NN were the top classifiers with MSEs in the range 0.073-0.074, and R2s in the range 0.856-0.859. This work demonstrates the effectiveness of guiding the feature pruning process by the values of the correlation coefficients in order to minimize the long training times of ML models while achieving good accuracies. © 2024 The Authors. Published by Elsevier B.V.

关键词： correlation coefficient feature pruning Internet of things IOT health intensive care unit dataset machine learning network security

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：