检索结果-内蒙古大学图书馆

Channel and Power Allocation for Multi-Cell NOMA Using Multi-Agent Deep Reinforcement Learning and Unsupervised Learning

引用

Sensors 2025年第9期25卷 2733-2733页

作者： Sun, Ming Zhong, Yihe He, Xiaoou Zhang, Jie College of Computer and Control Engineering Qiqihar University Qiqihar161006 China

Among the 5G and anticipated 6G technologies, non-orthogonal multiple access (NOMA) has attracted considerable attention due to its notable advantages in data throughput. Nevertheless, it is challenging to find the near-optimal allocation of the channel and power resources to maximize the performance of the multi-cell NOMA system. In addition, due to the complex and dynamically changing wireless communication environment and the lack of the near-optimal labels, conventional supervised learning methods cannot be directly applied. To address these challenges, this paper proposes a framework of MDRL-UL that integrates the multi-agent deep reinforcement learning with the unsupervised learning to allocate the channel and power resources in a near-optimal manner. In the framework, a multi-agent deep reinforcement learning neural network (MDRLNN) is proposed for channel allocation, while an attention-based unsupervised learning neural network (ULNN) is proposed for power allocation. Furthermore, the joint action (JA) derived from the MDRLNN for channel allocation is used as a representation to be fed into the ULNN for power allocation. In order to maximize the energy efficiency of the multi-cell NOMA system, the expectation of the energy efficiency is used to train both the MDRLNN and the ULNN. Simulation results indicate that the proposed MDRL-UL can achieve higher energy efficiency and transmission rates than other algorithms. © 2025 by the authors.

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

GenSUMO: GenAI-Creation of Critical Scenarios for Autonomous Vehicle Testing

GenSUMO: GenAI-Creation of Critical Scenarios for Autonomous...

引用

Consumer Communications and Networking Conference, CCNC IEEE

作者： Sergio Augusto Angelini Diego Gasco Claudio Casetti Department of Control and Computer Engineering Politecnico di Torino Turin Italy

ISBN: (数字)9798331508050

ISBN: (纸本)9798331508067

Recent advancements in autonomous vehicle research highlight the importance of Machine Learning (ML) models in tasks like motion planning, trajectory prediction, and emergency management. To support AI development, we propose a novel approach for generating on-demand datasets using the Simulator of Urban Mobility (SUMO) and a Generative Adversarial Network (GAN). Our method focuses on capturing critical events such as sudden pedestrian crossings, near-misses, and collisions, providing essential data to improve vehicle models' responses to emergency situations.

关键词： Pedestrians Machine learning Predictive models Generative adversarial networks Emergency services Data models Trajectory Planning Autonomous vehicles Testing

来源：评论

学校读者我要写书评

暂无评论

Security Offloading Scheduling and Caching Optimization Algorithm in UAV Edge Computing

引用

IEEE Systems Journal 2025年第1期19卷 96-106页

作者： Qiu, Jianli Kuang, Zhufang Huang, Zhenqi Lin, Siyu Central South University of Forestry and Technology College of Computer and Mathematics Changsha410004 China Beijing Jiaotong University School of Electronic and Information Engineering Beijing100044 China Beijing Jiaotong University State Key Laboratory of Rail Traffic Control and Safety Beijing100044 China

Mobile edge computing, a prospective wireless communication framework, can contribute to offload a large number of tasks to unmanned aerial vehicle (UAV) mobile edge servers. Besides, the demand for server computational resources increasingly ascends as the volume of processing tasks grows. However, in reality, many devices have similar computing tasks and require the same computing data. Therefore, servers can effectively reduce server computing latency and bandwidth costs by caching task data. This investigation explores task security offloading and data caching optimization strategies in scenarios with multiple interfering devices. With the goal of minimizing the total energy consumption, the UAV trajectories, transmission power, task offloading scheduling strategies, and caching decisions is jointly optimized. The corresponding optimization problem, which consists of mixed integer nonlinear programming problem, is formulated. To make this problem solved, the original problem is decomposed into three tiers, and an iterative algorithm named CDSFS which is based on the coordinate descent, successive convex approximation, and flow shop scheduling is proposed. Simulation results demonstrate the stability and superiority of the proposed algorithm. © 2007-2012 IEEE.

关键词： Nonlinear programming

来源：评论

学校读者我要写书评

暂无评论

MOHFL: Multi-Level One-Shot Hierarchical Federated Learning with Enhanced Model Aggregation over Non-IID Data

引用

IEEE Transactions on Network and Service Management 2025年

作者： Liu, Huili Ma, Yinglong Guo, Chenqi Liu, Xiaofeng Wang, Tingdong North China Electric Power University School of Control and Computer Engineering Beijing102206 China

Hierarchical federated learning (HFL) is a privacy-preserving distributed machine learning framework with a client-edge-cloud hierarchy, where multiple edge servers perform partial model aggregation to reduce costly communication with the cloud server. Nevertheless, most existing HFL methods require extensive iterative communication and public datasets, which not only increase communication overhead but also raise privacy and security concerns. Moreover, non-independent and identically distributed (non-IID) data among devices can significantly impact the accuracy of the global model in HFL. To address these challenges, we propose a multi-level one-shot HFL framework (MOHFL), which aims to improve the performance of the global model in a single communication round. Specifically, we employ conditional variational autoencoders (CVAEs) as local models and use the aggregated decoders to generate an IID training set for the global model, thereby mitigating the negative impact of non-IID data. We improve the performance of CVAEs under different levels of data heterogeneity through a dominant class-based data selection method. Subsequently, an edge aggregation scheme based on multi-teacher knowledge distillation and contrastive learning is proposed to aggregate the knowledge from local decoders to edge decoders. Extensive experiments on four real-world datasets demonstrate that MOHFL is very competitive against four state-of-the-art baselines under various settings. © 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

A Verifiable Trust-Based CP-ABE Access control Scheme for Cloud-Assisted Renewable Energy Systems

引用

computers, Materials & Continua 2025年第1期82卷 1211-1232页

作者： Jiyu Zhang Kehe Wu Ruomeng Yan Zheng Tian Yizhen Sun Yuxi Wu Yaogong Guo Department of Control and Computer Engineering North China Electric Power UniversityBeijing102206China State Grid Information&Communication Company of Hunan Electric Power Corporation Changsha410118China State Grid Hunan Electric Power Corporation Chenzhou Power Supply Branch Chenzhou423000China

Renewable Energy Systems(RES)provide a sustainable solution to climate warming and environmental pollution by enhancing stability and reliability through status acquisition and analysis on cloud platforms and intelligent processing on edge servers(ES).However,securely distributing encrypted data stored in the cloud to terminals that meet decryption requirements has become a prominent research ***,managing attributes,including addition,deletion,and modification,is a crucial issue in the access control scheme for *** address these security concerns,a trust-based ciphertext-policy attribute-based encryption(CP-ABE)device access control scheme is proposed for RES(TB-CP-ABE).This scheme effectivelymanages the distribution and control of encrypted data on the cloud through robust attribute key *** introducing trust management mechanisms and outsourced decryption technology,the ES system can effectively assess and manage the trust worthiness of terminal devices,ensuring that only trusted devices can participate in data exchange and access sensitive ***,the ES system dynamically evaluates trust scores to set decryption trust thresholds,thereby regulating device data access permissions and enhancing the system’s *** validate the security of the proposed TB-CP-ABE against chosen plaintext attacks,a comprehensive formal security analysis is conducted using the widely accepted random oraclemodel under the decisional q-Bilinear Diffie-Hellman Exponent(q-BDHE)***,comparative analysis with other schemes demonstrates that the TB-CP-ABE scheme cuts energy/communication costs by 43%,and scaleswell with rising terminals,maintaining average latency below 50ms,ensuring real-time service *** proposed scheme not only provides newinsights for the secure management of RES but also lays a foundation for future secure energy solutions.

关键词： Access control renewable energy systems(RES) ciphertext-policy attribute-based encryption(CP-ABE) security

来源：评论

学校读者我要写书评

暂无评论

TCIP: Transformed Colour Intensity Profiles analysis for fault detection in power line insulators

引用

Multimedia Tools and Applications 2025年第8期84卷 4503-4534页

作者： Tomaszewski, Michal Gasz, Rafal Kasana, Singara Singh Osuchowski, Jakub Singh, Simranjit Zator, Slawomir Department of Computer Science Faculty of Electrical Engineering Automatic Control and Informatics Opole University of Technology Prószkowska 76 Opole45-758 Poland Department of Computer Science and Information Technology Central University of Haryana Haryana Mahendergarh123031 India Department of Information Technology Dr B. R. Ambedkar National Institute of Technology Punjab Jalandhar India Department of Safety Engineering and Technical Systems Faculty of Production Engineering and Logistics Opole University of Technology Prószkowska 76 Opole45-758 Poland

In response to the escalating demand for electricity, the aging process and inherent failures in power lines have become unavoidable challenges in their operational integrity. This research addresses the imperative need for a robust method to detect damage in power insulator disc images, introducing a novel feature extraction technique grounded in color intensity profile transformation and ensemble learning. For the purposes of the research, a specific dataset was developed based on publicly available images of power insulators. The collection consists of 9232 images of insulator discs which included 2184 images of damaged discs. The discernible damages include breakages and burnouts. Two distinct analysis variants were employed. In the first variant (V1), two classification groups were selected: disc with damages and disc without damages. In the second variant (V2), three classification groups were selected: discs without damage, broken discs, and burnout discs. The described research was divided into two successive stages. In the first of them (S1), the transformed color intensity profiles for the prepared data set were built and various machine learning models for their classification were trained and evaluated. In the next stage (S2), previously developed, highest efficient machine learning models, were used to make the detection of various damaged types of overhead line insulator discs. For the first stage of research high-performance results and a high value of the F1 score parameters were obtained (F1 score = 0.999 for the first variant (S1V1) and F1 score = 0.992 for the second variant (S1V2). In the second stage of research and the first variant (S2V1), the highest result of the F1 score was 0.932 and for the second variant (S2V2) F1 score was 0.930. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Workflow as a Service Broker in Cloud Environment: A Systematic Literature Review

arXiv

引用

arXiv 2025年

作者： Abrishami, Saeid Zandi, Faridreza Momtaz Nourbakhsh, Alireza Department of Computer Engineering Faculty of Engineering Ferdowsi University of Mashhad Mashhad Iran Departement of Control and Computer Engineering Politecnico di Torino Torino Italy

Cloud computing has emerged as a promising platform for running scientific workflows across various domains. Scientists can take advantage of different cloud service models, such as serverful or serverless, to execute workflows based on their specific requirements, along with diverse pricing models like on-demand, reserved, or spot instances to reduce execution costs. However, the challenge of selecting appropriate resources and pricing models, coupled with the orchestration and scheduling of workflow tasks, creates significant complexity for users. To mitigate this burden, Workflow as a Service (WaaS) brokers have been introduced to facilitate workflow execution. In recent years, numerous studies have been published, either directly or indirectly related to this research area, highlighting the need for a comprehensive and systematic review of WaaS brokers to identify key trends and challenges in this field. In this paper, we conduct a Systematic Literature Review (SLR) on WaaS brokers within cloud environments. The SLR employs a thorough 3-tier strategy (database search, backward snowballing, and forward snowballing) to answer five research questions. A total of 74 high-quality articles, published in 43 prestigious venues, are analyzed to derive a taxonomy based on the architecture of WaaS brokers. The articles are classified and surveyed according to this taxonomy, and future research directions for the design and implementation of WaaS brokers are explored. This study provides valuable insights for researchers and developers, helping them identify major trends and issues in the field of WaaS brokers. © 2025, CC BY.

关键词： Cloud platforms

来源：评论

学校读者我要写书评

暂无评论

Large Language Models for topic analysis: insights from robotic accident narratives

引用

Procedia computer Science 2025年 253卷 1462-1472页

作者： Francesco Costantino Nicolò Sabetta Department of Computer Control and Management Engineering Sapienza University of Rome Via Ariosto 25 00185 Rome Italy

The integration of robotics in manufacturing has significantly enhanced productivity and safety by performing hazardous tasks. However, it also introduces new risks and accident profiles that require thorough analysis. This study explores the application of Large Language Models (LLMs) for topic modeling on accident narratives from the Occupational Safety and Health Administration (OSHA) database, focusing on incidents involving robots. Advanced Natural Language Processing (NLP) techniques, specifically LLMs, were employed to uncover prevalent themes, common risk factors, and patterns in these narratives. The methodology includes data collection, text processing, statistical analysis, and various topic modeling techniques. The analysis identifies six topics, and key themes including maintenance-related accidents, types of machinery involved, and specific body parts frequently affected. The findings highlight critical areas for improving safety measures, such as rigorous lockout/tagout procedures and enhanced training for maintenance personnel. This study provides a novel approach to analyze accident narratives.

关键词： Robot collaborative robot health safety NLP LLM topic modelling

来源：评论

学校读者我要写书评

暂无评论

Research on detecting internal trunk-boring insects in living trees based on BP-UNet

引用

Nondestructive Testing and Evaluation 2025年

作者： Zhou, Hongwei Xia, Bo Liu, Guoyang Zhou, Hongju College of Computer and Control Engineering Northeast Forestry University Harbin China Electronic Engineering College Heilongjiang University Harbin China

The invasion of trunk-boring insects halts tree growth and causes tree death, making it a key factor in forest destruction. To address the issue of low detection accuracy due to the small size and high concealment of trunk-boring insects, a combined driving algorithm based on BP-UNet is proposed. The BP algorithm exhibits high sensitivity to relative permittivity, enabling the detection of smaller insects. By learning the implicit geometric features from the BP results, UNet can accurately reconstruct the size and shape of the insects. To demonstrate the performance of the proposed algorithm, the BP-UNet is compared with the BP and UNet algorithms, and its generalisation ability in different scenarios is tested. Experimental results show that the BP-UNet algorithm has good generalisation and achieves a detection accuracy of over 80% even in the presence of noise, providing quantitative images of trunk-boring insects within 1 second. © 2025 Informa UK Limited, trading as Taylor & Francis Group.

关键词： Invertebrates

来源：评论

学校读者我要写书评

暂无评论

Cepstrum Lifting and Coherent Averaging-Based Instantaneous Frequency Estimation of Low-Amplitude Signal Components for Rotating Machinery

引用

IEEE Sensors Journal 2025年第10期25卷 17479-17490页

作者： Liu, Shaokang Liu, Shuai Wu, Yingjie Liu, Weiliang Liu, Changliang Luo, Zhihong North China Electric Power University Control and Computer Engineering Department Beijing102200 China Baoding Key Laboratory of State Detection and Optimization Regulation for Integrated Energy System Baoding071000 China Northeast Electric Power University Automation Engineering Department Jilin132000 China Inner Mongolia University of Technology Electric Power Department Hohhot010080 China

Instantaneous frequency (IF) serves multiple critical functions in the prognostic and health management of rotating machinery. The current IF estimation methodologies primarily concentrate on enhancing estimation accuracy at the algorithmic level. However, in practical industrial applications, the target signal component is often influenced by varying operating conditions, structural resonances, and background noise, causing it to become submerged or even disappear, ultimately resulting in the failure of conventional IF estimation techniques. To address these challenges, a novel IF estimation approach incorporating signal quality enhancement is developed. First, the complementary multisource vibration signal is processed by the proposed segmented cepstrum lifting (CL) method to overcome the submersion of the resonant frequency band on the target component. Then, the processed segmented signals are fused using coherent averaging (CA) technology to enhance the amplitude of the target component. Finally, leveraging the first-order approximation capability of the short-time fractional Fourier transform (STFrFT) for nonlinear frequency modulation modes, precise IF estimation is achieved. Verified by field data, the proposed method outperforms mainstream methods in both accuracy and applicability, highlighting its industrial application prospects. © 2001-2012 IEEE.

关键词： Frequency estimation

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：