检索结果-内蒙古大学图书馆

International Conference on computer Communications and Networks (ICCCN)

作者： Cheng Qian Mian Qian Kun Hua Hengshuo Liang Guobin Xu Wei Yu Dept. of Computer and Information Technology Hood College USA Dept. of Computer and Information Sciences Towson University Dept. of Electrical Engineering California Polytechnic State University USA Dept. of Computer Science Morgan State University USA

ISBN: (数字)9798350384611

ISBN: (纸本)9798350348439

The Internet of Vehicles (IoV), as one subset of the Internet of Things (IoT) in the smart transportation area, integrates vehicle networks with sensors and actuators. By connecting all sensors to the network, the IoV enables smart transportation (i.e., autonomous vehicles) and makes smart cities a reality. In smart transportation systems, roadside units (RSUs) capture all vehicle information and serve as gateways. However, smart transportation infrastructure has yet to mature in the current stage. RSUs are insufficient to support all vehicles. Meanwhile, the low computational capability of vehicles makes it challenging to recompute the driving route as the road environment changes. To address the problem of insufficient RSU coverage, one protocol called IEEE 802.11p enables vehicle-to-vehicle communication using relays. Nonetheless, data transfer among vehicles via relays is still time-consuming for a large-scale transportation network. To deal with the above issues, in this paper, we propose an IoV framework using digital twins (DTs) to digitize the IoV environment and assign nearby IoT gateways compatible with the RSU communication protocol. This framework lets DTs update the vehicle’s driving route based on real-time information. With a case study, we evaluate the efficacy of DT-assisted IoV based on communication latency and vehicle driving efficiency. Our evaluation results confirm that the proposed framework can efficiently enhance communication latency when the relay needs to pass through two or more vehicles and reduce travel time when vehicles receive updated route information at intersections.

关键词： Smart cities Vehicle driving Vehicular ad hoc networks Logic gates Routing protocols Smart transportation Digital twins

来源：评论

学校读者我要写书评

暂无评论

Multi-Agent Deep Reinforcement Learning for Access Point Activation Strategy in Cell-Free Massive MIMO Networks

Multi-Agent Deep Reinforcement Learning for Access Point Act...

引用

IEEE Conference on computer Communications Workshops, INFOCOM Wksps

作者： Li Sun Jing Hou Richard Chapman Department of Computer Science and Software Engineering Auburn University Auburn AL USA Department of Computer Science and Information System California State University San Marcos San Marcos CA USA

Cell-free massive MIMO network is a promising solution to support 5G communication. However, as an ultra-dense network (UDN), it suffers from two major issues: power consumption and scalability, which hinders it to be widely used in practice. The increasing and time-varying user demand, the dynamic propagation environment, and the huge amount of access points (APs) make it a challenging task to address these issues. For this purpose, we propose a distributed solution to solve the access point activation (APA) problem in cell-free massive MIMO networks to reduce power consumption considering dynamic user demand. We leverage the user-centric characteristic and design a multi-agent deep reinforcement learning (MADRL) algorithm by which each AP independently decides whether it needs to be switched on or off. The simulation results show that the MADRL outperforms the centralized and the random strategies. This research demonstrates the ability of distributed learning in solving the APA problem in cell-free massive MIMO networks.

关键词：

来源：评论

学校读者我要写书评

暂无评论

CtrlNeRF: The Generative Neural Radiation Fields for the Controllable Synthesis of High-fidelity 3D-Aware Images.

arXiv

引用

arXiv 2024年

作者： Liu, Jian Yu, Zhen School of Computer and Artificial Intelligence Zhengzhou University No.100 Science Avenue Henan Zhengzhou450001 China Department of Electrical and Computer Engineering California State Polytechnic University PomonaCA91768 United States

The neural radiance field (NERF) advocates learning the continuous representation of 3D geometry through a multilayer perceptron (MLP). By integrating this into a generative model, the generative neural radiance field (GRAF) is capable of producing images from random noise z without 3D supervision. In practice, the shape and appearance are modeled by zs and za, respectively, to manipulate them separately during inference. However, it is challenging to represent multiple scenes using a solitary MLP and precisely control the generation of 3D geometry in terms of shape and appearance. In this paper, we introduce a controllable generative model (i.e. CtrlNeRF) that uses a single MLP network to represent multiple scenes with shared weights. Consequently, we manipulated the shape and appearance codes to realize the controllable generation of high-fidelity images with 3D consistency. Moreover, the model enables the synthesis of novel views that do not exist in the training sets via camera pose alteration and feature interpolation. Extensive experiments were conducted to demonstrate its superiority in 3D-aware image generation compared to its counterparts. © 2024, CC BY.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Ctrlnerf: A Generative Neural Radiation Field for the Controllable Synthesis of High-Fidelity 3d-Aware Images

SSRN

引用

SSRN 2024年

The neural radiance field (NERF) advocates the implicit representation of 3D geometry using a multilayer perceptron (MLP). Novel views were synthesized by querying 5D coordinates along the camera ray and projecting the output color and density into an image using volume rendering. By incorporating this into a generative model, the generative neural radiance field (GRAF) can generate images from random noise $z$ without 3D supervision. In practice, the shape and appearance are modeled by $z_{s}$ and $z_{a}$, respectively, to manipulate them separately during inference. However, it is challenging to represent multiple scenes using a solitary MLP and to precisely control the generation of 3D geometry in terms of shape and appearance. In this study, we propose a generative model ($i.e.$ \textbf{CtrlNeRF}) that utilizes a single MLP network to represent multiple scenes with shared weights. Consequently, we manipulated the input of the shape and appearance codes to realize controllable generation of high-fidelity images with 3D consistency. Moreover, the model enables the synthesis of novel views that do not exist in the training sets via camera-pose alteration and feature interpolation. Extensive experiments were conducted to demonstrate its superiority in 3D-aware image generation compared to its counterparts. © 2024, The Authors. All rights reserved.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Ctrlnerf: The Generative Neural Radiation Fields for the Controllable Synthesis of High-Fidelity 3d-Aware Images

SSRN

引用

SSRN 2024年

The neural radiance field (NERF) advocates learning the continuous representation of 3D geometry via multilayer perceptron (MLP). By incorporating this into a generative model, the generative neural radiance field (GRAF) can generate images from random noise z without 3D supervision. In practice, the shape and appearance are modeled by zs and za, respectively, to manipulate them separately during the inference. However, it is challenging to represent multiple scenes using a solitary MLP and to precisely control the generation of 3D geometry in terms of shape and appearance. In this study, we propose a controllable generative model (i.e. CtrlNeRF) that utilizes a single MLP network to represent multiple scenes with shared weights. Consequently, we manipulated the shape and appearance codes to realize controllable generation of high-fidelity images with 3D consistency. Moreover, the model enables the synthesis of novel views that do not exist in the training sets via camera-pose alteration and feature interpolation. Extensive experiments were conducted to demonstrate its superiority in 3D-aware image generation compared with its counterparts. © 2024, The Authors. All rights reserved.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

computer-Vision-Based Non-Contact Paste Concentration Measurement*

Computer-Vision-Based Non-Contact Paste Concentration Measur...

引用

Automation in Manufacturing, Transportation and Logistics (ICaMaL), International Conference on

作者： Tailin Liang Zhaolin Yuan State Key Laboratory of Software Development Environment School of Computer Science and Engineering Beihang University Beijing China Department of Industrial and Systems Engineering Research Institute of Advanced Manufacturing The Hong Kong Polytechnic University Hong Kong China

ISBN: (数字)9798350378658

ISBN: (纸本)9798350378665

For monitoring the paste concentration, existing techniques, such as ultrasonic concentration meters and neutron meters, suffer from radiation hazards and low precision in high concentrations. This paper proposes a novel non-contact concentration measurement method based on deep learning. With the dataset collected by preparing the paste with different concentrations and taking photographs from the paste surface, a convolutional neural network is trained to extract the features from images and predict the concentration. The experiments indicate that the measurement accuracy is close to 88.79 % for the manual stirring paste dataset and 91.42 % for the automatic stirring paste dataset, which are sufficient in industrial applications. As a substitution of traditional concentration measurement instruments, the proposed vision-based concentration measurement method is non-contact and its accuracy is sufficient in industrial applications.

关键词： Meters Deep learning Accuracy Ultrasonic variables measurement Transportation Manuals Feature extraction Surface texture Convolutional neural networks Surface treatment

来源：评论

学校读者我要写书评

暂无评论

A Stylometric Dataset for Bengali Poems 6

A Stylometric Dataset for Bengali Poems

引用

6th International Conference on Natural Language Processing and Information Retrieval, NLPIR 2022

作者： Shuhan, Mirza Kamrul Bashar Dey, Rupasree Saha, Sourav Anjum, Md Shafa Ul Zaman, Tarannum Shaila Software Research and Engineering BKash Limited Dhaka Bangladesh Computer Science Oregon State University CorvallisOR United States Business Administration Shahjalal University of Science and Technology Sylhet Bangladesh Progoti Systems Limited Dhaka Bangladesh Suny Polytechnic Institute UticaNY United States

ISBN: (纸本)9781450397629

Poetry is a form of literature that conveys feelings using different styles, aesthetics, and rhythms. The Bengali language has an enriched collection of poems. Every poet has an individual style of expressing their thoughts and emotions. However, stylometric research in this branch of the Bengali language is still in its early stage of development. In this paper, we have presented a stylometric dataset, which has 6,070 poems of 137 poets stored in the textual format. To the best of our knowledge, this is the first stylometric dataset for Bengali poems which will add an extra dimension to the expanding research arena of the Bengali language. To explore the usability of this dataset, we developed poem genre classifiers using deep learning that can classify these poems. Performance analysis of some deep learning classifiers has been presented in addition to classification. The classifiers include GRU and CNN. Among these two, GRU showed better performance by 91.48 in terms of the F1-score. The dataset will be publicly available at https://***/shuhanmirza/Bengali-Poem-Dataset after publishing this article. © 2022 ACM.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

Optimizing ASCON Permutation in Multi-Clock Domains with Chisel: Resource Efficiency and Critical Path Reduction

Optimizing ASCON Permutation in Multi-Clock Domains with Chi...

引用

Proceedings of the IEEE Dallas Circuits and Systems Workshop (DCAS)

作者： Mohamed El-Hadedy Russell Hua Kazutomo Yoshii Wen-Mei Hwu Department of Electrical and Computer Engineering California State Polytechnic University Pomona USA Coordinated Science Laboratory University of Illinois at Urbana-Champaign USA Argonne National Laboratory Lemont Illinois USA

ISBN: (数字)9798350349535

ISBN: (纸本)9798350349542

This study delves into optimizing the ASCON cryptographic protocol, specifically its core permutation function critical for the sponge-construction mechanism, within Field Programmable Gate Array (FPGA) platforms to enhance Internet of Things (IoT) device security. By implementing a novel approach that marries multi-clock domain strategies with pipelining, we substantially elevate the permutation function's efficiency and operational frequency. Our FPGA-based ASCON IP showcases an operational frequency of 233.3MHz, leveraging the distinct advantages of multi-clock domains to mitigate inherent hardware resource constraints and achieve reasonable performance despite Application Specific Integrated Circuits (ASIC)s traditionally reaching higher frequencies up to 633MHz. Notably, the architecture's innovative division into two stages-first, integrating the add constant and substitution and second, concentrating on the diffusion layer-enables optimized processing speeds by allowing each stage to function at its maximum effective clock rate, thereby enhancing overall throughput. The first stage achieves a remarkable frequency of up to 465 MHz, utilizing 210 Look-Up Table (LUTs), 362 Flip-Flops (FF)s, and 0.5 Block RAM (BRAM), while the second stage operates up to 233 MHz with 848 LUTs and 623 FFs. This strategic configuration, supported by a double-rating buffer, ensures seamless data flow between stages, significantly improving the permutation process's efficiency. The proposed design, meticulously developed in Chisel, exemplifies how integrating high-level programming agility with precise hardware control can spur innovation in cryptographic hardware design, propelling the development process forward without compromising on performance or security. Through this advanced approach, we set a new benchmark for implementing lightweight, high-performance cryptographic solutions, aiming to secure the proliferating IoT ecosystem effectively.

关键词： Performance evaluation Frequency-domain analysis Process control Hardware Table lookup Cryptography Internet of Things

来源：评论

学校读者我要写书评

暂无评论

SENTINEL: Scalable Edge Network for Telepresence and Integrated AI

SENTINEL: Scalable Edge Network for Telepresence and Integra...

引用

Telepresence, IEEE Conference on

作者： Mohamed El-Hadedy James Ung David Kim Wen-Mei Hwu Yutao He Benny Cheng Department of Electrical and Computer Engineering California State Polytechnic University Pomona USA Coordinated Science Laboratory University of Illinois at Urbana-Champaign USA University of California Los Angeles USA Corona Division Naval Surface Warfare Center USA

ISBN: (数字)9798331528249

ISBN: (纸本)9798331528256

This paper presents a research platform centered on an RC car with a hybrid system utilizing TurningPi and DeskPi6C, with multiple Raspberry Pi Compute Module 4 nodes for edge computing. The platform is designed to facilitate telepresence applications by integrating advanced computer vision (YOLO) and natural language processing (NLP) models (BERT), managed by a backend server. The system’s NLP module achieves command processing latencies ranging from 1000 to 3000 μs in ideal scenarios, enabling near-instantaneous responses in telepresence scenarios. The YOLOv5s model, deployed on the edge, supports real-time visual feedback with an impressive inference speed of approximately 10 FPS. The system’s scalability demonstrated through Kubernetes, ensures that YOLO inference speed significantly improves as more pods are added, up to the platform’s 10-node capacity, thereby highlighting its potential for handling the increased workload. The system’s efficiency and reliability are underscored by its power consumption and resource utilization metrics, which are critical for sustained telepresence operations. The central server maintains stable power usage at around 57.8 W, while the DeskPi Super 6C and Turing Pi nodes peak at 36.0 W and 20.7 W, respectively, under load. This ensures robust performance in remote environments, reassuring the audience of its suitability for sustained telepresence operations. This platform showcases the potential of integrating Kubernetes with edge computing to enhance telepresence applications, offering real-time AI processing and scalable operations. The research provides valuable insights into developing versatile, efficient systems for telepresence, enabling more prosperous and responsive remote interactions.

关键词： YOLO Visualization Telepresence Computational modeling Scalability Natural language processing Real-time systems Servers Artificial intelligence Edge computing

来源：评论

学校读者我要写书评

暂无评论

A Neural Collaborative Filtering Model for Adverse Drug Reaction Prediction 2

A Neural Collaborative Filtering Model for Adverse Drug Reac...

引用

2nd International Conference on Data Analytics, Computing and Artificial Intelligence, ICDACAI 2023

作者： Xiong, Zetong Du, Zihao Rong, Xi Yang, Yuting Zhang, Xiaoyi Zhongnan University of Economic and Law School of Information and Safety Engineering Wuhan430073 China McMaster University Department of Computing and Software HamiltonL8S 4L8 Canada Tufts University School of Engineering MedfordMA02155 United States The Ohio State University Department of Computer Science and Engineering Columbus43212 United States University of California Department of Mathematics San Diego921222 United States

ISBN: (纸本)9798350339765

As adverse drug reactions (ADRs) can cause serious consequences to medication treatment, identifying and predicting adverse drug reactions (ADRs) play a crucial role in drug effects and drug use safety. In this paper, we reviewed existing machine learning methods in ADR prediction and proposed a neural collaborative filtering model (NCF) for the prediction of monopharmacy ADRs. NCF is based on dimension reduction by matrix factorization (MF) and combined with an MLP model to enhance prediction performance by handling complex non-linear relationships using MLN. Our NCF model optimizes the MF model by replacing dot product in the MF model with MLP, integrating drug and ADR embeddings from monopharmacy ADR benchmark data and drug-related chemical, physical and biological descriptors in a two-layer deep neural network. The model was tested by using 10-fold cross-validation. On 10-fold cross-validation, the resulting AUC, AUPR, and running time indicated a better performance and higher efficiency on drug-ADR prediction achieved by applying our NCF model. © 2023 IEEE.

关键词： Collaborative filtering

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：