检索结果-内蒙古大学图书馆

2nd International Joint Conference on Information and Communication Engineering, JCICE 2023

作者： Ren, Shuidi Xu, Xiaodong Han, Shujun Beijing University of Posts and Telecommunications State Key Laboratory of Networking and Switching Technology Beijing China

ISBN: (纸本)9798350325768

In the future Internet of Intelligent Things, when multiple devices have too much computation task at one moment, device-collaborative computation offloading helps them to reduce the task pressure while it may cause more users to suffer from task overload due to computation task delivery. Finding the critical conditions that lead to the all-user task overload, it helps us deploy the communication network to guarantee its stability. Therefore, in this paper we study the process of computation task delivery among the mobile edge networks (MENs) and its impact. We utilize the spreading threshold model in propagation dynamics theory to characterize the device-collaborative computation offloading. The edge-based compartmental theory is applied to describe the proposed model, establish the propagation dynamics equations, and derive the saddle points. Through analysis and simulation, we find that the network status that all users suffer from data overload can be avoided by controlling the proportion of users who choose device-collaborative computation offloading and the network heterogeneity. © 2023 IEEE.

关键词： Computation offloading

来源：评论

学校读者我要写书评

暂无评论

Maximizing Energy Efficiency for LEO Satellite Systems With Massive MIMO-NOMA 9

Maximizing Energy Efficiency for LEO Satellite Systems With ...

引用

9th International Conference on Computer and Communications, ICCC 2023

作者： Li, Lihua Liu, Xin Beijing University of Posts and Telecommunications State Key Laboratory of Networking and Switching Technology Beijing China

ISBN: (纸本)9798350317251

In this paper, we consider the application of Non-Orthogonal Multiple Access (NOMA) in Low Earth Orbit (LEO) satellite communication systems (SCS) with hybrid beamforming structure to support Massive Machine Communication (mMTC). Satellites are typically powered by solar energy, leading to significant energy costs, therefore, we propose a resource allocation scheme aimed at maximizing the energy efficiency of the system. Specifically, we firstly group users based on their channel correlations using the agglomerative nesting (AGNES) algorithm. Subsequently, a joint hybrid beamforming and power allocation problem is formulated to maximize the energy efficiency of the system. To address this non-convex problem with high-dimensional variables, we first obtain the power allocation solution under fixed hybrid beamforming using fractional Programming (FP) and Successive Convex Approximation (SCA) methods. Then, given fixed analog beamforming, we utilize the approximate zero-forcing method to design the digital beamforming. Finally, a Boundary Compressed-Particle Swarm Optimization (BC-PCO) algorithm is used to obtain the analog beamforming. The simulation results demonstrate that the proposed algorithm exhibits significant superiority in terms of energy efficiency. © 2023 IEEE.

关键词： Energy efficiency

来源：评论

学校读者我要写书评

暂无评论

Service Migration in Mobile Edge Computing Based on 5G Network User Plane

Service Migration in Mobile Edge Computing Based on 5G Net...

引用

International Conference on Big Data Intelligence and Computing, DataCom 2022

作者： Hui, Liu QiBo, Sun State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunications Beijing China

ISBN: (纸本)9789819922321

The development of mobile edge computing under 5G network provides basic technical support for low latency and high reliability and also poses new challenges to service migration. 5G provides an uplink classifier (UL CL) function in data transmission, which makes it possible to build a new data transmission path. This paper proposes a new service migration decision model based on the 5G network user plane. Based on the new service migration model, this paper uses Markov decision process to construct service migration decision problem, and proposes an online service migration algorithm based on reinforcement learning to learn the optimal migration strategy. Finally, we compare the proposed algorithm with other benchmark algorithms. The results show that the online algorithm is close to the global performance, superior to the existing benchmark algorithm in terms of problem formulation proposed in our paper. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： 5G mobile communication systems

来源：评论

学校读者我要写书评

暂无评论

FRACTAL: Data-aware Clustering and Communication Optimization for Decentralized Federated Learning

引用

IEEE Transactions on Big Data 2024年 1-16页

作者： Ma, Qianpiao Liu, Jianchun Xu, Hongli Jia, Qingmin Xie, Renchao Purple Mountain Laboratories Nanjing Jiangsu China School of Computer Science and Technology University of Science and Technology of China Hefei Anhui China State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunications Beijing China

Decentralized federated learning (DFL) is a promising technique to enable distributed machine learning over edge nodes without relying on a centralized parameter server. However, existing DFL network topologies, such as fully connected, partially connected, or lower-tier hierarchical topology often struggle to effectively address the unique challenges presented by edge networks, including edge heterogeneity, communication resource constraint, and data Non-IID. In order to tackle these challenges, we propose a data-aware clustering algorithm, called FRACTAL, to construct a multi-tier hierarchical topology in a bottomup manner taking into consideration both data distribution and communication efficiency for DFL. We theoretically explore the quantitative relationship between the convergence bound of multi-tier FL and the data distribution among each-tier servers. To further improve communication efficiency and address edge heterogeneity, we deploy a time-sharing communication scheduling algorithm within each fractal unit (the basic structure in FRACTAL consisting of multiple nodes and an aggregator), called magic mirror method (MMM), to determine the optimal order of model distributing and uploading for nodes. We conduct extensive experiments on the classical models and datasets to evaluate the performance of FRACTAL, and the results show that FRACTAL can significantly accelerate the DFL model training by 48.6%- 72.3% compared with the state-of-the-art solutions. IEEE

关键词： Network topology

来源：评论

学校读者我要写书评

暂无评论

Multi-Relational Knowledge Enhanced Contrastive Learning for Event Extraction 9

Multi-Relational Knowledge Enhanced Contrastive Learning for...

引用

9th International Conference on Computer and Communications, ICCC 2023

作者： Liu, Yixuan Shuang, Kai Beijing University of Posts and Telecommunications State Key Laboratory of Networking and Switching Technology Beijing China

ISBN: (纸本)9798350317251

Event extraction (EE) is a fundamental task under the field of natural language processing. Most of the existing EE methods perform supervised fine-tuning, training downstream with the help of the ability to pre-train language models (PLMs), limited labeled data is difficult to satisfy the training needs of PLMs and faces problems such as limited generalization. Contrastive learning, as a self-supervised training paradigm, improves the representational power of PLMs by acquiring a wider range of knowledge from large-scale unlabeled corpora than is implied by the data labels. However, existing contrastive learning methods mainly focus on the sentence-level, while the development at the token-level represented by EE tasks is still in its early stages, resulting in noisy representations of tokens and a lack of detailed use of event knowledge, which makes the performance bottleneck of the PLMs after contrastive learning when fine-tuning downstream EE tasks. To this end, we propose MRKE-CLEE, a multi-relational knowledge enhanced contrastive learning for event extraction. MRKE-CLEE first accomplishes the coreference resolution and AMR parsing as supported tasks to obtain multiple event-relationship knowledge pseudo-labels from a large-scale unlabeled corpus through the auxiliary model, then uses in-sentence knowledge guided contrast to filter noisy contrastive pairs to make the resulting token representations more distinct, and using event type knowledge guided contrast to aggregate token representations with the same event type. With the combination of multi-relational event knowledge, using contrastive learning to improve the performance of event detection (ED) and event argument extraction (EAE). Experiments demonstrate that our MRKE-CLEE achieves a new state-of-the-art performance on the ACE2005 and MAVEN datasets and introduces new attempts to use token-level contrastive learning in EE tasks. © 2023 IEEE.

关键词： Extraction

来源：评论

学校读者我要写书评

暂无评论

Stable 5G Time Domain Resource Configuration for Synchronous Timing Services via Lyapunov Aided DRL 36

Stable 5G Time Domain Resource Configuration for Synchronous...

引用

36th IEEE/IFIP Network Operations and Management Symposium, NOMS 2023

作者： Zhou, Yanbo Feng, Lei Xie, Kunyi Zhou, Fanqin Li, Wenjing Yu, Peng Beijing University of Posts and Telecommunications The State Key Laboratory of Networking and Switching Technology Beijing China

ISBN: (纸本)9781665477161

The high-precision clock synchronization is pursued with the consideration of the balance for time-domain resource utilization, when 5G technologies are expected to carry the timing services. Firstly, a clock synchronization model is established in the case that the system exists the observed value loss. The probability distribution of the loss of the observed value is evaluated by a newly proposed delay deterministic confidence method. Then, the error boundness of the clock synchronization is investigated by the Kalman filter algorithm, and the optimization problem for 5G time-domain resource configuration is formulated with guaranteeing the precision of clock synchronization;Finally, the proposed optimization problem is solved by dueling double deep Q-learning-based Lyapunov optimization. The experimental results verify the effectiveness and superiority of the proposed method in terms of the joint optimization of clock synchronization error covariance and throughput. © 2023 IEEE.

关键词： Optimization

来源：评论

学校读者我要写书评

暂无评论

Long-tailed Detection Based on Multi-Expert Aggregation

Long-tailed Detection Based on Multi-Expert Aggregation

引用

2023 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2023

作者： Luo, Yang Lin, Rongheng State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunications Beijing China

ISBN: (纸本)9798350337020

Long-tailed data distribution is a common phenomenon observed in nature. However, this distribution poses challenges for neural network training, as it performs well on samples with head categories, but not so well on tail categories - precisely what requires attention in certain scenes. Although existing methods involve re-sampling data or designing loss functions, they often fail to adapt well to changing data distributions. In this article, we propose a long-tailed detection approach based on multi-expert aggregation (MEA). This method employs multi-expert networks to learn about head, medium, and tail categories from a fixed long-tailed distribution. Different data are modelled using different loss designs, and the results of multi-expert networks are aggregated to achieve good accuracy for both head and tail categories. The effectiveness of this method has been verified on LVIS. © 2023 IEEE.

关键词： deep learning long-tailed recognition multi-expert learning object detection

来源：评论

学校读者我要写书评

暂无评论

BlindSpotEliminator: Collaborative Point Cloud Perception in Cellular-V2X Networks

BlindSpotEliminator: Collaborative Point Cloud Perception in...

引用

2023 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2023

作者： Chen, Ziyue Luo, Guiyang Shao, Congzhang Yuan, Quan Li, Jinglin Beijing University of Posts and Telecommunications State Key Laboratory of Networking and Switching Technology Beijing China

ISBN: (纸本)9798350337020

Multi-agent collaborative perception depends on sharing sensory information to improve perception accuracy and robustness, as well as to extend coverage. However, most collaborative perception methods ignore the limitations of communication networks, such as limited bandwidth and the possibility of wireless conflicts. To fill this gap, this paper proposes BlindSpotEliminator, a conflict-free scheduler over the cellular-V2X networks for supporting practical collaborative point cloud perception to eliminate blind spots. BlindSpotEliminator first identifies the blind spots for each vehicle, then lists the corresponding conflict relationships based on the distribution of the blind spots and communication conflicts, and finally designs an optimized point cloud data transmission strategy to eliminate the blind spots of each vehicle. Extensive experiments show that compared with greedy algorithm and random methods, BlindSpotEliminator achieves better efficiency, i.e., transmitting 20% more point cloud data. © 2023 IEEE.

关键词： Cellular V2X collaborative perception optimization strategy point cloud data transmission

来源：评论

学校读者我要写书评

暂无评论

Design and Development of a Log Management System Based on Cloud Native Architecture 9

Design and Development of a Log Management System Based on C...

引用

9th International Conference on Systems and Informatics, ICSAI 2023

作者： Sun, Yuchen Chen, Yanpiao Zhao, Haotian Peng, Shan Beijing University of Posts and Telecommunications State Key Laboratory of Networking and Switching Technology Beijing China

ISBN: (纸本)9798350383706

With the rapidly development and wide application of cloud computing and service computing, the use of cloud native architecture has become a mainstream trend in software development and deployment. In this booming background, logs have become the key information for each application to record its operation process throughout its life cycle, while existing log management systems are usually designed for specific production environments, which are not able to effectively meet the growing and diversified scenarios. The existing log management systems are usually designed for specific production environments and cannot effectively meet the needs of the growing number of diverse scenarios. Therefore, how to effectively handle log information from multiple sources becomes a key issue. In this paper, we design and develop a log management platform based on cloud-native architecture, which provides log collection, transmission, storage, and system management functions, and at the same time can be oriented to the cloud-native environment and other third-party applications for collection, providing a powerful tool for developers and operation and maintenance personnel. In addition, we designed an anomaly detection system using large language models for log parsing and used this system for anomaly detection in platform logs. Finally, the effectiveness of the system was verified by commonly used log data. The log management platform based on cloud-native architecture passed the test and has been utilized in our real production. © 2023 IEEE.

关键词： Software design

来源：评论

学校读者我要写书评

暂无评论

A Federated Learning Approach for Net Load Forecasting in Microgrids. 24

A Federated Learning Approach for Net Load Forecasting in Mi...

引用

24th Asia-Pacific Network Operations and Management Symposium, APNOMS 2023

作者： Qin, Zihan Shao, Sujie Guo, Shaoyong Qiu, Xuesong State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunications Beijing China

ISBN: (纸本)9788995004395

Microgrid is a small-scale power grid composed of various distributed energy resources, and net load forecasting (NLF) plays significant role in its energy generation planning. Due to the dispersed distribution of nodes within the microgrid and disparate user data, completing the net load forecasting task becomes challenging. In order to efficiently forecast the net load while ensuring data security, this paper proposes an approach for NLF based on federated learning (FL). Firstly, a FL framework for NLF in microgrids is constructed to complete the NLF task with long short-term memory neural network (LSTM) model and node selection (NS) in aggregation. Secondly, considering factors such as the data precision and quantity, an accuracy optimization problem model is proposed. Then, the problem is formulated as a Markov decision process (MDP), and a NS algorithm based on Asynchronous Advantage Actor-Critic (A3C) is designed for selecting a node set to optimize model aggregation. The simulation demonstrate that the proposed NS method based on A3C improves the performance of FL, resulting in better convergence of LSTM network and ensuring the scalability of the cluster in microgrid. Copyright 2023 KICS.

关键词： Long short-term memory

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：