检索结果-内蒙古大学图书馆

13th International Conference on Computer Engineering and Networks, CENet 2023

作者： Zhao, Yikun Zhang, Jinli Zhou, Fanqin Li, Wenjing Feng, Lei State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunications Beijing China

ISBN: (纸本)9789819992386

Millimeter wave (mmWave) aerial base stations offer distinct advantages such as enhanced flexibility and mobility compared to ground base station communication. Additionally, the utilization of mmWave technology enables the provision of high-speed, directional services to ground users while mitigating interference with existing ground networks. To establish a good communication link between mmWave aerial base stations and terrestrial mobile users, it is essential for base stations to form directional beam to cover terrestrial users. However, the movement of users and the change of environmental factors will lead to the misalignment between the mmWave beam formed by the base stations and the users, resulting in the instability of the communication link, which will seriously affect the quality of communication service. Therefore, this paper proposes a deep learning-based beam tracking scheme for mmWave aerial base stations. Simulation results show that our proposed schemes achieve higher beamforming gain and require less beam training overhead compared to existing conventional and deep learning-based approaches. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd 2024.

关键词： Base stations

来源：评论

学校读者我要写书评

暂无评论

Optimizing Dual-Mode UAV-Assisted Remote IoT Data Collection with Decentralized DRL in 6G-Enabled Space-Air-Ground Networks

Optimizing Dual-Mode UAV-Assisted Remote IoT Data Collection...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

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

ISBN: (纸本)9798350368741

As sixth-generation (6G) networks extend to remote regions, efficient data collection from Internet of Things (IoT) devices faces challenges such as limited infrastructure, diverse latency requirements, and energy constraints. To address these, we propose a dual-mode data transmission strategy: (1) a UAV-satellite network for real-time delay-sensitive data, where wireless power transfer (WPT) is introduced to alleviate energy depletion caused by frequent data transmission, and a deadline τmax is set in each timeslot to maintain data freshness. (2) a carry-store mode for delay-tolerant data. Both aim to minimize energy consumption while maximizing data collection. Specifically, we develop a time scheduling policy to allocate data and energy transmission time for delay-sensitive data within τmax, and propose a novel centralized control, distributed execution framework using decentralized deep reinforcement learning (DRL), termed DRL-Schedule, designed to optimize UAV navigation and ensure energy-efficient, timely data collection. Simulation results demonstrate that DRL-Schedule achieves an average improvement of approximately 11.2% and 12.8% over the best baseline PPO in energy efficiency, when varying different number of UAVs and IoTs, respectively, and notably surpassing GA, Greedy and Random. © 2025 IEEE.

关键词： 6G Deep reinforcement learning Energy-efficiency UAV-LEO integrated data collection WPT

来源：评论

学校读者我要写书评

暂无评论

Coordinating Communication and Computing for Wireless VR in Open Radio Access Networks

引用

IEEE Transactions on Mobile Computing 2025年第7期24卷 6075-6089页

作者： Li, Hongtao Chen, Ziqi Guo, Fengxian Li, Nan Sun, Yaohua Peng, Mugen Liu, Yuanwei Beijing University of Posts and Telecommunications State Key Laboratory of Networking and Switching Technology Beijing100876 China China Mobile Research Institute Beijing100053 China Hong Kong

Driven by diverse applications, radio access networks (RAN) are expected to embrace built-in computing and intelligence, forming a versatile wireless computing platform that closely integrates communication and computing. To fully unleash the potential of such a synergistic system, it is essential to coordinate communication and computing with intelligence unlocked by the radio intelligent controllers (RICs) in O-RAN. Building on the groundwork established by existing theoretical studies and simulations, we develop a platform that can emulate the events in the real-world system in more detail, bringing theoretical works closer to practical implementation. In this paper, we first introduce ns-GP-O-RAN, a software simulation platform developed over ns-3, enabling communication, computation task processing, large-scale data collection, and testing of system-level orchestration policies through user-level control. Taking virtual reality (VR) as an example, we formulate the computation offloading problem and develop a prediction-based computation offloading xAPP, which contains a prediction phase to predict users’ end-to-end (E2E) performance with the deep neural network and a system-level decision-making phase for global orchestration with the differential evolution algorithm. We evaluate the system capacity and E2E latency over the developed ns-GP-O-RAN, which is more effective than existing approaches. © 2002-2012 IEEE.

关键词： Radio access networks

来源：评论

学校读者我要写书评

暂无评论

LEP: Leveraging Local Entropy Pruning for Sparsity in Large Language Models

LEP: Leveraging Local Entropy Pruning for Sparsity in Large ...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Chen, Yuli Cheng, Bo Zhang, Yingying Zhang, Shuhao Wu, Zhixuan Meng, Fanshen State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunications China

ISBN: (纸本)9798350368741

The application of Large Language Models (LLMs) is rapidly expanding in fields such as natural language processing and computer vision. However, due to the enormous number of model parameters, while their emergent capabilities enhance performance, they also incur significant computational and storage costs, particularly during inference. This makes it particularly challenging to deploy large models in resource-constrained environments. To address this issue, we propose an innovative pruning method based on local entropy, aimed at improving model performance while reducing computational costs. Our approach leverages a small amount of calibration data and calculates the importance of weights based on the local entropy of weights and input activations, applying it to both unstructured and semi-structured pruning without the need for complex weight updates or retraining. Experimental results demonstrate that our method shows superior performance across models of various scales. Specifically, at a sparsity rate of 50%, our method significantly reduces model perplexity without the need for weight updates and outperforms existing pruning techniques such as SparseGPT and Wanda in zero-shot tasks. The code is available at: this https URL. © 2025 IEEE.

关键词： entropy large language models model compression unstructed pruning

来源：评论

学校读者我要写书评

暂无评论

Delay and Energy Consumption Oriented UAV Inspection Business Collaboration Computing Mechanism in Edge Computing Based Electric Power IoT

引用

Chinese Journal of Electronics 2023年第1期32卷 13-25页

作者： SHAO Sujie LI Yi GUO Shaoyong WANG Chenhui CHEN Xingyu QIU Xuesong State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunications Blockchain Research Department China Electronics Standardization Institute

With the development of Internet of things(IoT) technology and smart grid infrastructure,edge computing has become an effective solution to meet the delay requirements of the electric power IoT. Due to the limitation of battery capacity and data transmission mode of IoT terminals, the business collaboration computing must consider the energy consumption of the terminals. Since delay and energy consumption are the optimization goals of two co-directional changes, it is difficult to find a business collaboration computing mechanism that simultaneously minimizes delay and energy *** paper takes the unmanned aerial vehicle(UAV) inspection business scenario in the electric power IoT based on edge computing as the representative, and proposes a two-stage business collaboration computing mechanism including resources allocation and task allocation to optimize the business delay and energy consumption of UAV by decoupling the complex correlation between resource allocation and task allocation. A steepest descent resource allocation algorithm is proposed. On the basis of resource allocation, an improved multiobjective evolutionary algorithm based on decomposition by dynamically adjusting the size of neighborhood and the cross distribution index is proposed as a task allocation algorithm to minimize energy consumption and business delay. Simulation results show that our algorithms can respectively reduce the business delay and energy consumption by more than6.4% and 9.5% compared with other algorithms.

关键词： Energy consumption Heuristic algorithms Collaboration Autonomous aerial vehicles Delays Power systems Resource management

来源：评论

学校读者我要写书评

暂无评论

Associative Tasks Computing Offloading Scheme in Internet of Medical Things with Deep Reinforcement Learning

引用

China Communications 2024年第4期21卷 38-52页

作者： Jiang Fan Qin Junwei Liu Lei Tian Hui School of Communications and Information Engineering Xi’an University of Posts and TelecommunicationsXi’an 710121China State Key Laboratory of Networking and Switching Technology Beijing University of Posts and TelecommunicationsBeijing 100876China

The Internet of Medical Things(Io MT) is regarded as a critical technology for intelligent healthcare in the foreseeable 6G era. Nevertheless, due to the limited computing power capability of edge devices and task-related coupling relationships, Io MT faces unprecedented challenges. Considering the associative connections among tasks, this paper proposes a computing offloading policy for multiple-user devices(UDs) considering device-to-device(D2D) communication and a multi-access edge computing(MEC)technique under the scenario of Io MT. Specifically,to minimize the total delay and energy consumption concerning the requirement of Io MT, we first analyze and model the detailed local execution, MEC execution, D2D execution, and associated tasks offloading exchange model. Consequently, the associated tasks’ offloading scheme of multi-UDs is formulated as a mixed-integer nonconvex optimization problem. Considering the advantages of deep reinforcement learning(DRL) in processing tasks related to coupling relationships, a Double DQN based associative tasks computing offloading(DDATO) algorithm is then proposed to obtain the optimal solution, which can make the best offloading decision under the condition that tasks of UDs are associative. Furthermore, to reduce the complexity of the DDATO algorithm, the cacheaided procedure is intentionally introduced before the data training process. This avoids redundant offloading and computing procedures concerning tasks that previously have already been cached by other UDs. In addition, we use a dynamic ε-greedy strategy in the action selection section of the algorithm, thus preventing the algorithm from falling into a locally optimal solution. Simulation results demonstrate that compared with other existing methods for associative task models concerning different structures in the Io MT network, the proposed algorithm can lower the total cost more effectively and efficiently while also providing a tradeoff between delay and energy co

关键词： associative tasks cache-aided procedure double deep Q-network Internet of Medical Things(IoMT) multi-access edge computing(MEC)

来源：评论

学校读者我要写书评

暂无评论

Collaborative Multiple Access And Energy-Efficient Resource Allocation in Distributed Maritime Wireless Networks

引用

China Communications 2022年第4期19卷 137-153页

作者： Xueyan Cao Hongming Zhang Mugen Peng State Key Laboratory of Networking and Switching Technology Beijing University of Posts and TelecommunicationsBeijing 100876China

With the rapid increasing of maritime activities, maritime wireless networks(MWNs) with high reliability, high energy efficiency, and low delay are required. However, the centralized networking with fixed resource scheduling is not suitable for MWNs due to the special environment. In this paper,we introduce the collaborative relay communication in distributed MWNs to improve the link reliability, and propose an orthogonal time-frequency resource block reservation based multiple access(RRMA) scheme for both one-hop direct link and two-hop collaborative relay link to reduce the interference. To further improve the network performance, we formulate an energy efficiency(EE) maximization resource allocation problem and solve it by an iterative algorithm based on the Dinkelbach method. Finally, numerical results are provided to investigate the proposed RRMA scheme and resource allocation algorithm, showing that the low outage probability and transmission delay can be attained by the proposed RRMA scheme. Moreover,the proposed resource allocation algorithm is capable of achieving high EE in distributed MWNs.

关键词： maritime wireless networks distributed multiple access resource allocation energy efficiency maximization

来源：评论

学校读者我要写书评

暂无评论

Edge Intelligence Assisted Resource Management for Satellite Communication

引用

China Communications 2022年第8期19卷 31-40页

作者： Yaohua Sun Mugen Peng State Key Laboratory of Networking and Switching Technology Beijing University of Posts and TelecommunicationsBeijing 100876China

Satellite communication has been seen as a vital part of the sixth generation communication,which greatly extends network *** satellite communication,resource management is a key problem attracting many research ***,previous study mainly focuses on throughput improvement via power allocation and spectrum assignment and the proposed approaches are mostly model-based and dedicated to specific problem ***,with the trend of edge intelligence,complex resource management problems can be efficiently resolved in a model-free *** this paper,a joint beam activation,user-beam association and time resource allocation approach is *** core idea is using stochastic learning at the ground station to identify active user-link beams to meet user rate *** addition,the convergence,optimality and complexity of our proposal are rigorously *** simulation,it is shown that the rate goal of most of the users can be met and meanwhile satellite energy is saved owing to much less active beams.

关键词： satellite communication resource management stochastic learning

来源：评论

学校读者我要写书评

暂无评论

A Double-Timescale Reinforcement Learning Based Cloud-Edge Collaborative Framework for Decomposable Intelligent Services in Industrial Internet of Things

引用

China Communications 2024年第10期21卷 181-199页

作者： Zhang Qiuyang Wang Ying Wang Xue School of Information and Communication Engineering Beijing University of Posts and TelecommunicationsBeijing 100876China State Key Laboratory of Networking and Switching Technology(Beijing University of Posts and Telecommunications) Beijing 100876China

With the proportion of intelligent services in the industrial internet of things(IIoT)rising rapidly,its data dependency and decomposability increase the difficulty of scheduling computing *** this paper,we propose an intelligent service computing *** the framework,we take the long-term rewards of its important participants,edge service providers,as the optimization goal,which is related to service delay and computing *** the different update frequencies of data deployment and service offloading,double-timescale reinforcement learning is utilized in the *** the small-scale strategy,the frequent concurrency of services and the difference in service time lead to the fuzzy relationship between reward and *** solve the fuzzy reward problem,a reward mapping-based reinforcement learning(RMRL)algorithm is proposed,which enables the agent to learn the relationship between reward and action more *** large time scale strategy adopts the improved Monte Carlo tree search(MCTS)algorithm to improve the learning *** simulation results show that the strategy is superior to popular reinforcement learning algorithms such as double Q-learning(DDQN)and dueling Q-learning(dueling-DQN)in learning speed,and the reward is also increased by 14%.

关键词： computing service edge intelligence industrial internet of things(IIoT) reinforcement learning(RL)

来源：评论

学校读者我要写书评

暂无评论

ADC-DL:Communication-Efficient Distributed Learning with Hierarchical Clustering and Adaptive Dataset Condensation

引用

China Communications 2022年第12期19卷 73-85页

作者： Zhipeng Gao Yan Yang Chen Zhao Zijia Mo State Key Laboratory of Networking and Switching Technology Beijing University of Posts and TelecommunicationsBeijing 100876China

The rapid growth of modern mobile devices leads to a large number of distributed data,which is extremely valuable for learning ***,model training by collecting all these original data to a centralized cloud server is not applicable due to data privacy and communication costs concerns,hindering artificial intelligence from empowering mobile ***,these data are not identically and independently distributed(Non-IID)caused by their different context,which will deteriorate the performance of the *** address these issues,we propose a novel Distributed Learning algorithm based on hierarchical clustering and Adaptive Dataset Condensation,named ADC-DL,which learns a shared model by collecting the synthetic samples generated on each *** tackle the heterogeneity of data distribution,we propose an entropy topsis comprehensive tiering model for hierarchical clustering,which distinguishes clients in terms of their data ***,synthetic dummy samples are generated based on the hierarchical structure utilizing adaptive dataset *** procedure of dataset condensation can be adjusted adaptively according to the tier of the *** experiments demonstrate that the performance of our ADC-DL is more outstanding in prediction accuracy and communication costs compared with existing algorithms.

关键词： distributed learning Non-IID data partition hierarchical clustering adaptive dataset condensation

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：