检索结果-内蒙古大学图书馆

IEEE Transactions on Sustainable Computing 2024年第3期10卷 534-546页

作者： Tian, Jiwei Shen, Chao Wang, Buhong Ren, Chao Xia, Xiaofang Dong, Runze Cheng, Tianhao Ministry of Education Key Laboratory for Intelligent Networks and Network Security School of Cyber Science and Engineering Xi'an Jiaotong University Xi'an China Air Traffic Control and Navigation College Air Force Engineering University Xi'an China Information and Navigation College Air Force Engineering University Xi'an China School of Electrical and Electronic Engineering Nanyang Technological University Singapore School of Computer Science and Technology Xidian University Xi'an China

Although conventional false data injection attacks can circumvent the detection of bad data detection (BDD) in sustainable power grid cyber physical systems, they are easily detected by well-trained deep learning-based detectors. Still, state estimation models with deep leaning-based detectors are not secure due to the vulnerabilities and fragility of deep learning models. Using the related laws of conventional false data injection attacks and adversarial sample attacks, this paper proposes the targEted adVersarial fAlse Data injEction (EVADE) strategy to explore targeted adversarial false data injection attacks for state estimation in Smart Grid. The proposed EVADE attack strategy selects key state variables based on adversarial saliency maps to improve the attack efficiency and perturbs as few state variables as possible to reduce the attack cost. In this way, the EVADE attack strategy can bypass the detection of BDD and neural attack detection (NAD) methods (that is, maintaining deep stealthy) with a high success rate and achieve the attack target simultaneously. Experimental results demonstrate the effectiveness of the proposed strategy, posing serious and pressing concerns for sustainable cyber physical power system security. © 2016 IEEE.

关键词： Cyber attacks

来源：评论

学校读者我要写书评

暂无评论

OptoGPT: A foundation model for inverse design in optical multilayer thin film structures

引用

Opto-Electronic Advances 2024年第7期7卷 4-16页

作者： Taigao Ma Haozhu Wang L.Jay Guo Department of Physics University of MichiganAnn ArborMichigan 48109USA Department of Electrical Engineering and Computer Science University of MichiganAnn ArborMichigan 48109USA

Optical multilayer thin film structures have been widely used in numerous photonic ***,existing inverse design methods have many drawbacks because they either fail to quickly adapt to different design targets,or are difficult to suit for different types of structures,e.g.,designing for different materials at each *** methods also cannot accommodate versatile design situations under different angles and *** addition,how to benefit practical fabrications and manufacturing has not been extensively considered *** this work,we introduce OptoGPT(Opto Generative Pretrained Transformer),a decoder-only transformer,to solve all these drawbacks and issues simultaneously.

关键词： multilayer thin film structure inverse design foundation models deep learning structural color

来源：评论

学校读者我要写书评

暂无评论

NAaLOSS: Rethinking the Objective of Speech Enhancement 33

NAaLOSS: Rethinking the Objective of Speech Enhancement

引用

33rd IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2023

作者： Ho, Kuan-Hsun Yu, En-Lun Hung, Jeih-Weih Chen, Berlin National Taiwan Normal University Department of Computer Science and Information Engineering Taipei Taiwan National Chi Nan University Department of Electrical Engineering Nantou Taiwan

ISBN: (纸本)9798350324112

Reducing noise interference is crucial for automatic speech recognition (ASR) in a real-world scenario. However, most single-channel speech enhancement (SE) generates 'processing artifacts' that negatively affect ASR performance. Hence, in this study, we suggest a Noise- and Artifacts-aware loss function, NAaLoss, to ameliorate the influence of artifacts from a novel perspective. NAaLoss considers the loss of estimation, de-artifact, and noise ignorance, enabling the learned SE to individually model speech, artifacts, and noise. We examine two SE models (simple/advanced) learned with NAaLoss under various input scenarios (clean/noisy) using two configurations of the ASR system (with/without noise robustness). Experiments reveal that NAaLoss significantly improves the ASR performance of most setups while preserving the quality of SE toward perception and intelligibility. Furthermore, we visualize artifacts through waveforms and spectrograms, and explain their impact on ASR. © 2023 IEEE.

关键词： Speech enhancement

来源：评论

学校读者我要写书评

暂无评论

ViT-Reg: Regression-Focused Hardware-Aware Fine-Tuning for ViT on tinyML Platforms

IEEE Design and Test

引用

IEEE Design and Test 2024年

作者： Shaharear, Md Ragib Mazumder, Arnab Neelim Mohsenin, Tinoosh Johns Hopkins University Department of Electrical & Computer Engineering United States University of Maryland Department of Computer Science & Electrical Engineering Baltimore County United States

Vision Transformers (ViTs) have demonstrated significant improvements in image classification tasks. However, deploying them on resource-constrained tinyML platforms presents considerable challenges due to their high computational demands and dynamic power consumption. Current methods rely heavily on computationally intensive architecture search techniques to identify optimal configurations, which are not well-suited for tinyML devices. This paper introduces ViT-Reg, a regression-based hardware-aware fine-tuning approach that identifies suitable ViT architectures for tinyML platforms. The proposed method enables efficient exploration of the configuration space, drastically reducing the computational overhead typically associated with architecture searches. ViT-Reg is hardware-aware, utilizing polynomial regression to narrow the search space while treating accuracy as a constraint. In experiments conducted on the CIFAR-10 and Tiny-ImageNet datasets, ViT-Reg deployed on Nvidia Jetson Nano achieved a 55.6% and 37.4% reduction in dynamic power consumption, along with a 65% and 60% improvement in energy efficiency compared to baseline ViT models. Finally, ViT-Reg provides an 8× improvement in energy efficiency relative to recent hardware implementations of the VGG model. © 2024 IEEE.

关键词： real-time and energy efficient deployment Regression tinyML Hardware Vision Transformer

来源：评论

学校读者我要写书评

暂无评论

Detecting the Number of Bite Prehension of Ggrazing Cows in an Extensive System Using an Audio Recording Method 8

Detecting the Number of Bite Prehension of Ggrazing Cows in ...

引用

8th International Conference of Yearly Reports on Informatics, Mathematics, and engineering, ICYRIME 2023

作者： Avanzato, Roberta Avondo, Marcella Beritelli, Francesco Franco, Francesco Di Tumino, Serena Department of Electrical Electronic and Computer Engineering University of Catania Catania Italy Department of Agricultural Food and Environmental Science University of Catania Catania Italy

In the context of cattle farming, understanding the feeding behavior of animals is essential to ensure their welfare and maximize productivity. However, monitoring and interpret- ing the acoustic signals associated with grazing, particularly the sound events related to grass intake, pose a significant challenge. This study proposes an innovative method based on 1D convolutional neural networks to automatically classify such sound events during grazing. The approach was developed using a balanced dataset composed of 322 prehension samples and 1000 non-prehension samples, extracted from audio recordings of grazing cattle in real conditions. The results obtained show a high accuracy of 100% during the testing and validation phases of the model. However, there is concern about overfitting of the model due to the limited size of the dataset used. Consequently, future expansion of the dataset is suggested by collecting a larger and more diverse number of audio recordings to improve the generalization and robustness of the model in real cattle farming contexts. © 2023 Copyright for this paper by its authors.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Compressive Spectral Video via Wavelength-Dynamic Coded Aperture

Compressive Spectral Video via Wavelength-Dynamic Coded Aper...

引用

2023 Photonics North, PN 2023

作者： Norato, D. Marquez, M. Jerez, A. Arguello, H. Meza, P. Universidad Industrial de Santander Department of Computer Science Bucaramanga Colombia Universidad de la Frontera Department of Electrical Engineering Temuco Chile

ISBN: (纸本)9798350326734

Compressive spectral video (CSV) has emerged as a novel research topic in computer vision, having powerful implications in retrieving a four-dimensional tensor from a reduced set of two-dimensional compressed measurements. The advances in CSV architectures have shown that the recovery of four-dimensional data is possible, but the quality is still far from the requirements of real applications. To overcome these limitations, this work analyses the parameter design of a novel dynamic spectro-temporal encoding methodology based on a digital micromirror device sync with a tunable filter. © 2023 IEEE.

关键词： Image compression

来源：评论

学校读者我要写书评

暂无评论

Blockchain-based Secure UAV-assisted Battlefield Operation underlying 5G 14

Blockchain-based Secure UAV-assisted Battlefield Operation u...

引用

14th International Conference on Information and Knowledge Technology, IKT 2023

作者： Pancholi, Dhruvi Jadav, Nilesh Kumar Tanwar, Sudeep Garg, Deepak Zanjani, S. Mohammadali Institute of Technology Nirma University Department of Computer Science and Engineering Ahmedabad 382481 India SR University Distinguished Professor School of Computer Science and Artificial Intelligence Warangal 506371 India Islamic Azad University Najafabad Branch Department of Electrical Engineering Najafabad Iran

ISBN: (纸本)9798350349412

The use of unmanned aerial vehicles (UAVs) has significantly increased during the past few years. UAVs have a wide range of applications, but their employment in the defense sector has grown significantly in recent years as a result of their exceptional effectiveness in reducing casualties and enabling the accomplishment of high-profile as well as time-sensitive operations. UAVs are helpful for numerous defense-related operations, such as border patrol, surveillance, emergency management, damage assessment, delivery of medical supplies and armaments, and many others. However, because wireless communication is open, UAVs can be used to undermine the effectiveness of military operations. In this paper, we developed a blockchain-based framework to mitigate security threats from UAVs operating in a combat scenario. The MATLAB virtual environment is used to simulate a battlefield scenario, including UAVs, sensors, enemies, and friendly nodes. The UAV scenario also includes an enemy that continuously manipulates the UAVs to endanger battlefield operations. The suggested framework employs machine learning algorithms to find malicious UAVs. Further, the proposed framework uses blockchain technology to increase security and lessen data tampering attempts. A smart contract (for data validation) is deployed on an IPFS-based public blockchain. In addition, 5G technology is used, which makes it possible for UAVs and the control station to communicate effectively with lower packet error rates and latency. The proposed framework is evaluated by considering evaluation metrics, such as 5G packet error rate, latency, and transaction and execution costs for the blockchain. © 2023 IEEE.

关键词： Blockchain

来源：评论

学校读者我要写书评

暂无评论

Deep Matrix Factorization for Multi-view Clustering Using Density-Based Preprocessing 1

引用

2nd International Conference on Pan-African Intelligence and Smart Systems, PAAISS 2022

作者： Ahiaklo-Kuz, Raphael K. M. Asiedu, Charles Jnr. Ngatched, Telex M. N. Woungang, Isaac Huzhou University Zhejiang China Department of Electrical and Computer Engineering Memorial University St. John’sNL Canada Department of Computer Science Toronto Metropolitan University TorontoON Canada

ISBN: (数字)9783031252716

ISBN: (纸本)9783031252709

Using algebraic tools to advance autonomous data mining is difficult because the best clustering techniques do not scale with the dataset’s modulo/cardinality, especially in this era of data surge. This paper suggested a pre-processing strategy based on density to compact primary data through a collection of representative points while maintaining the original data’s density and distribution. This study used data compression to preserve the dataset’s density distribution and features. Most multiview clustering approaches use a single-layer formulation. Since the mapping between the new presentation and the primary data contains intricate hierarchical details, it’s more efficient to investigate hidden structures hierarchically. The emphasis is on learning the complementary information to solve the clustering problem. This study presented a deep matrix factorization framework that made use of multi-view clustering (MVC). Semi-nonnegative matrix factorization (NMF) learns multi-view data hierarchies layer-by-layer. To effectively mine shared information from each view, the output layer must be identical. Graph regularizers combine deep structure layer representations to maintain geometry across views. Benchmark results show the model’s effectiveness. © 2023, ICST Institute for computer sciences, Social Informatics and Telecommunications engineering.

关键词： Matrix factorization

来源：评论

学校读者我要写书评

暂无评论

IoT-Assisted AI Framework for Preserving Biodiversity: Monitoring Rare Trees in Dynamic Forest Ecosystems

IoT-Assisted AI Framework for Preserving Biodiversity: Monit...

引用

2025 International Conference on Multi-Agent Systems for Collaborative Intelligence, ICMSCI 2025

作者： Rajan, K. Palanivel Raj, J. Relin Francis Krishnan, R. Santhana Balamuralikrishnan, G. Vidyasree, G. Jothi, J. Nirmal Psna College of Engineering and Technology Department of Electrical and Electronics Engineering Dindigul India Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Department of Electronics and Communication Engineering Chennai India Scad College of Engineering and Technology Department of Electronics and Communication Engineering Cheranmahadevi India Velammal College of Engineering and Technology Department of Computer Science and Engineering Madurai India Sri Sai Ranganathan Engineering College Department of Computer Science and Engineering Coimbatore India Vel Tech Rangarajan Dr. Sagunthala R&d Institute of Science and Technology Department of Electronics and Communication Engineering Avadi India

ISBN: (纸本)9798331509828

The preservation of biodiversity, particularly the monitoring of rare tree species in dynamic forest ecosystems, presents significant challenges due to the complexities in detecting and analyzing forest changes. This research introduces a novel IoT -assisted AI framework, termed FoLiA-Net (Forest Lightweight Adaptive Network), designed to improve the efficiency and accuracy of monitoring rare trees. The proposed algorithm combines IoT sensor data with machine learning techniques to adaptively detect, classify, and track rare tree species across large and dynamically changing forest areas. The framework utilizes a lightweight neural network structure to ensure real-time processing while minimizing computational resources, making it suitable for deployment in resource-constrained environments. Simulation experiments were conducted to evaluate the performance of FoLiA-Net in comparison with existing algorithms, including Faster R-CNN (F-R-CNN), Random Forest Classifier (RFC), and Gradient Boosting Machines (GBM). The evaluation was based on several key performance metrics such as accuracy, latency and detection efficiency. The results revealed that FoLiA-Net outperformed the existing algorithms in terms of accuracy and computational efficiency, demonstrating a notable improvement in the monitoring of rare tree species in diverse forest ecosystems. The proposed framework is particularly effective in large-scale forest monitoring, providing a valuable tool for biodiversity conservation efforts © 2025 IEEE.

关键词： Biodiversity Forest Ecosystems Machine Learning Rare Trees Simulation Sustainability

来源：评论

学校读者我要写书评

暂无评论

Next-Gen Telehealth Ecosystems Using IoT for Virtual Consultations and Support 10

Next-Gen Telehealth Ecosystems Using IoT for Virtual Consult...

引用

10th International Conference on Communication and Signal Processing, ICCSP 2024

作者： Kathija Nasreen, A. Priya, S. Gupta, Nukala Sujata Shoba, L.K. Mohankumar, N. Arunkumar, S. Rmk College of Engineering and Technology Department of Computer Science and Engineering Tamil Nadu Chennai India Deemed To Be University Department of Electrical and Electronics Engineering Tamil Nadu Chennai India Chaitanya Bharathi Institute of Technology Department of Computer Engineering and Technology Telangana Hyderabad India Srm Institute of Science and Technology Department of Computing Technologies Chenagalpattu Chennai Tamil Nadu Kattankulathur India Symbiosis Institute of Technology Maharashtra Nagpur India Adhiparasakthi Engineering College Department of Computer Applications Tamil Nadu Melmaruvathur India

ISBN: (纸本)9798350353068

The research discussed focuses on improving virtual medical consultations and support using Internet of Things (IoT) technologies in telehealth services. It is predicted that IoT devices can assist with creating a new telehealth environment where patients and doctors can communicate easily for remote monitoring, diagnosis, and treatment. The main objective of this unique approach is to expand access to healthcare, especially in poor and rural areas. Connected medical equipment, smart sensors, and wearable technology all play an important role in this system. These IoT components enable gathering real-time and complete patient data, helping doctors make better, more individualized choices and care plans. Data safety, privacy, and system compatibility are some areas where this integration hits a snag. Sensitive patient data and the integrity of the IoT-enabled systems need robust safety measures. These IoT improvements in telemedicine have had a revolutionary effect. Proactive interventions, continuous health monitoring, and individualized care are ways patients might benefit from improved medical care. Virtual consultations are increasingly used in the medical field, and a more connected and data-driven framework for the patient experience is emerging. © 2024 IEEE.

关键词： Internet of things

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：