检索结果-内蒙古大学图书馆

Efficient Parameter Synchronization for Peer-to-Peer Distributed Learning With Selective multicast

IEEE TRANSACTIONS ON SERVICES COMPUTING 2025年第1期18卷 156-168页

作者： Luo, Shouxi Fan, Pingzhi Li, Ke Xing, Huanlai Luo, Long Yu, Hongfang Southwest Jiaotong Univ Sch Comp & Artificial Intelligence Chengdu 611756 Peoples R China Southwest Jiaotong Univ CSNMT Int Cooperat Res Ctr Key Lab Informat Coding & Transmiss Chengdu 611756 Peoples R China Univ Elect Sci & Technol China Sch Informat & Commun Engn Chengdu 611731 Peoples R China

Recent advances in distributed machine learning show theoretically and empirically that, for many models, provided that workers will eventually participate in the synchronizations, i) the training still converges, even if only p workers take part in each round of synchronization, and ii) a larger p generally leads to a faster rate of convergence. These findings shed light on eliminating the bottleneck effects of parameter synchronization in large-scale data-parallel distributed training and have motivated several optimization designs. In this paper, we focus on optimizing the parameter synchronization for peer-to-peer distributed learning, where workers broadcast or multicast their updated parameters to others for synchronization, and propose SELMCAST, a suite of expressive and efficient multicast receiver selection algorithms, to achieve the goal. Compared with the state-of-the-art (SOTA) design, which randomly selects exactly p receivers for each worker's multicast in a bandwidth-agnostic way, SELMCAST chooses receivers based on the global view of their available bandwidth and loads, yielding two advantages, i.e., accelerated parameter synchronization for higher utilization of computing resources and enlarged average p values for faster convergence. Comprehensive evaluations show that SELMCAST is efficient for both peer-to-peer Bulk Synchronous Parallel (BSP) and Stale Synchronous Parallel (SSP) distributed training, outperforming the SOTA solution significantly.

关键词： Training Synchronization Receivers Peer-to-peer computing Convergence Distance learning Computer aided instruction Bandwidth Optimization multicast algorithms Distributed learning parameter synchronization receiver selection

来源：评论

学校读者我要写书评

暂无评论

Fast Beamforming Power Minimization in NOMA Unicast and multicast Systems With Rate Guarantees

引用

IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS 2025年第4期24卷 3242-3257页

作者： Gau, Rung-Hung Chiu, Hsuan-Ming Natl Yang Ming Chiao Tung Univ Inst Commun Engn Hsinchu 300 Taiwan

We propose fast beamforming algorithms for minimizing the power consumption of wireless communications while assuring unicast and multicast data rates in NOMA wireless networks. Specifically, we put forward the greedy backward substitution algorithm for unicast beamforming design and the Gram-Schmidt rotation algorithm for multicast beamforming design. The proposed approach adopts zero-forcing techniques, orthogonalization and rotation but does not have to solve complicated convex optimization problems. In addition, we derive novel analytical results for the studied problem. Simulation results indicate that the proposed approach outperforms a number of alternative schemes in the literature in terms of beamforming power consumption and computational complexity.

关键词： NOMA Array signal processing Vectors Wireless communication Unicast multicast algorithms Resource management Base stations Power demand Communication system security Beamforming design non-orthogonal multiple access data rate guarantees multicast non-convex optimization linear algebra

来源：评论

学校读者我要写书评

暂无评论

Cache-Aware Cooperative multicast Beamforming in Dynamic Satellite-Terrestrial Networks

引用

IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY 2025年第1期74卷 1433-1445页

作者： Yuan, Shuo Sun, Yaohua Peng, Mugen Beijing Univ Posts & Telecommun State Key Lab Networking & Switching Technol Beijing 100876 Peoples R China

With the burgeoning demand for data-intensive services, satellite-terrestrial networks (STNs) face increasing backhaul link congestion, deteriorating user quality of service (QoS), and escalating power consumption. Cache-aided STNs are acknowledged as a promising paradigm for accelerating content delivery to users and alleviating the load of backhaul links. However, the dynamic nature of low earth orbit (LEO) satellites and the complex interference among satellite beams and terrestrial base stations pose challenges in effectively managing limited edge resources. To address these issues, this paper proposes a method for dynamically scheduling caching and communication resources, aiming to reduce network costs in terms of transmission power consumption and backhaul traffic, while meeting user QoS demands and resource constraints. We formulate a mixed timescale problem to jointly optimize cache placement, LEO satellite beam direction, and cooperative multicast beamforming among satellite beams and base stations. To tackle this intricate problem, we propose a two-stage solution framework, where the primary problem is decoupled into a short-term content delivery subproblem and a long-term cache placement subproblem. The former subproblem is solved by designing an alternating optimization approach with whale optimization and successive convex approximation methods according to the cache placement state, while cache content in STNs is updated using an iterative algorithm that utilizes historical information. Simulation results demonstrate the effectiveness of our proposed algorithms, showcasing their convergence and significantly reducing transmission power consumption and backhaul traffic by up to 52%.

关键词： Array signal processing Satellite broadcasting Base stations Backhaul networks Satellites Optimization multicast algorithms Power demand Space-air-ground integrated networks Low earth orbit satellites Beam direction control cache placement multicast beamforming satellite-terrestrial networks (STNs)

来源：评论

学校读者我要写书评

暂无评论

Mobile Tile-Based 360^° Video multicast With Cybersickness Alleviation

引用

IEEE TRANSACTIONS ON MOBILE COMPUTING 2025年第4期24卷 3423-3440页

作者： Lin, Chiao-Wen Yang, De-Nian Liao, Wanjiun Natl Taiwan Univ Dept Elect Engn Taipei City Taiwan Acad Sinica Inst Informat Sci Res Ctr Informat Technol Innovat Taipei City Taiwan

Virtual reality (VR) imaging is 360 degrees, which requires a large bandwidth for video transmission. To address this challenge, tile-based streaming has been proposed to deliver only the focused part of the video instead of the entire one. However, the impact of cybersickness, akin to motion sickness, on tile selection in VR has not been explored. In this paper, we investigate Multi-user Tile Streaming with Cybersickness Control (MTSCC) in an adaptive 360(degrees) video streaming system with multicast and cybersickness alleviation. We propose a novel m(2)-competitive online algorithm that utilizes Individual Sickness Indicator (ISI) and Bitrate Restriction Indicator (BRI) to evaluate user cybersickness tendency and network bandwidth efficiency. Moreover, we introduce the Video Loss Indicator (VLI) and Quality Variance Indicator (QVI) to assess video quality loss and quality difference between tiles. We also propose a multi-armed bandit (MAB) algorithm with confidence bound-based reward (video quality) and cost (cybersickness) estimation. The algorithm learns the weighting factor of each user's cost to slow down cybersickness accumulation for users with high cybersickness tendencies. We prove that the algorithm converges to an optimal solution over time. According to simulation with real network settings, our proposed algorithms outperform baselines in terms of video quality and cybersickness accumulation.

关键词： Cybersickness Streaming media Video recording Quality assessment Bandwidth multicast algorithms Optical flow Mobile computing Costs User experience VR 360(degrees) video tile-based streaming cybersickness multicast

来源：评论

学校读者我要写书评

暂无评论

multicast Scheduling Over Multiple Channels: A Distribution-Embedding Deep Reinforcement Learning Method

引用

IEEE TRANSACTIONS ON MOBILE COMPUTING 2025年第5期24卷 3502-3519页

作者： Li, Ran Huang, Chuan Qin, Xiaoqi Yang, Dong Nie, Xinyao Chinese Univ Hong Kong Shenzhen Future Network Intelligence Inst Shenzhen 518172 Peoples R China Chinese Univ Hong Kong Dept Informat Engn Hong Kong Peoples R China Chinese Univ Hong Kong Sch Sci & Engn Shenzhen Future Network Intelligence Inst Shenzhen 518172 Peoples R China Chinese Univ Hong Kong Guangdong Prov Key Lab Future Networks Intelligenc Shenzhen 518172 Peoples R China Beijing Univ Posts & Telecommun State Key Lab Networking & Switching Technol Beijing 100876 Peoples R China Beijing Jiaotong Univ Sch Elect & informat Engn Beijing 100044 Peoples R China SF Technol Shenzhen 518052 Peoples R China

multicasting is an efficient technique for simultaneously transmitting common messages from the base station (BS) to multiple mobile users (MUs). multicast scheduling over multiple channels, which aims to jointly minimize the energy consumption of the BS and the latency of serving asynchronized requests from the MUs, is formulated as an infinite-horizon Markov decision process (MDP) problem with a large discrete action space, multiple time-varying constraints, and multiple time-invariant constraints. To address these challenges, this paper proposes a novel distribution-embedding multi-agent proximal policy optimization (DE-MAPPO) algorithm, which consists of one modified MAPPO and one distribution-embedding module. The former one handles the large discrete action space and time-varying constraints by modifying the structure of the actor networks and the training kernel of the conventional MAPPO;and the latter one iteratively adjusts the action distribution to satisfy the time-invariant constraints. Moreover, a performance upper bound of the considered MDP is derived by solving a two-step optimization problem. Finally, numerical results demonstrate that our proposed algorithm outperforms the existing ones in terms of applicability, effectiveness, and robustness, and achieves comparable performance to the derived upper bound.

关键词： multicast algorithms Energy consumption Optimal scheduling Streaming media Quality of service Cellular networks NOMA multicast communication Millimeter wave communication Upper bound multicast large discrete action space time-varying constraints time-invariant constraints distribution-embedding multi-agent proximal policy optimization (DE-MAPPO)

来源：评论

学校读者我要写书评

暂无评论

Enhancing Multi-Connected XR Device Performance With Soft Combining and Joint HARQ Processing in Cooperative multicast

IEEE OPEN JOURNAL OF THE COMMUNICATIONS SOCIETY

引用

IEEE OPEN JOURNAL OF THE COMMUNICATIONS SOCIETY 2025年 6卷 2438-2452页

作者： Ahsen, Muhammad Yanakiev, Boyan Rosa, Claudio Adeogun, Ramoni Aalborg Univ Dept Elect Syst DK-9220 Aalborg Denmark Nokia Stand DK-9220 Aalborg Denmark

In this work, we investigate multi-connectivity for Extended Reality (XR) devices, which have stringent requirements in terms of delay, data rate and reliability. This approach employs one direct connection and one indirect connection via a 5th generation-Advanced (5G-A) tethering device. We investigate the Selection Combining Scheme (SCS) and Selection/Soft Combining Scheme (SSCS) between the tethering device and the XR device under point-to-multipoint (PTM) multicast transmission in downlink (DL) in a spatially consistent channel model. The SSCS is particularly effective in enhancing the XR device performance, achieving the maximum potential gain of either selection or soft combining, while limiting cooperative data exchange. This is done through various limited fraction soft bits relaying schemes, including modified ones from existing literature and newly proposed code block (CB) based indexing schemes. Additionally, we propose algorithms for joint processing of hybrid automatic repeat request (HARQ) feedback (FB) in 5G multicast sessions with cooperative users to enhance spectral efficiency by enabling the network to benefit from the cooperation of the users. Our link-level simulations demonstrate that multi-connectivity with joint HARQ enables XR devices to achieve the same spectral efficiency as with unicast but with a 0.8 dB and 4.5 dB lower average signal-to-noise ratio (SNR) with SCS and SSCS, respectively, in non-line-of-sight (NLOS) scenarios. In line-of-sight (LOS) scenarios, multi-connectivity yields a gain of 3 dB with SSCS, though no significant gain is observed with SCS.

关键词： Diversity reception 5G mobile communication Decoding Reliability Receivers multicast algorithms Collaboration Base stations Spectral efficiency Quality of service XR wireless tethering cooperative communication MBS HARQ tethered subnetwork soft combining

来源：评论

学校读者我要写书评

暂无评论

Deep Reinforcement Learning Based Intelligent Resource Allocation in Hybrid Vehicle Scenario

引用

IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY 2025年第3期74卷 4656-4668页

作者： Lou, Chengkai Hou, Fen Li, Bo Ding, Hongwei Univ Macau State Key Lab IoT Smart City Macau 999078 Peoples R China Univ Macau Dept ECE Macau 999078 Peoples R China Yunnan Univ Sch Informat Sci & Engn Kunming 650106 Peoples R China

In recent years, there has been rapid development in vehicular networks and autonomous driving. While vehicles of various intelligence levels are becoming more common on the road, most research overlooks the data distribution across different vehicles in multicast scenarios. Our aim is to allow different kinds of vehicles to receive the needed content in a multicast scenario and to fulfill certain freshness requirements. Although deep reinforcement learning (DRL) has been widely used to address this issue, it suffers from slow training convergence and unstable performance. Hence, this study proposes combining DRL algorithms with behavior cloning and action mask, leveraging prior knowledge and expert algorithms to enhance performance. Finally, the freshness of the data content is ensured for all kinds of vehicles and effective data transmission is achieved. The simulation results indicate a significant improvement in the training efficiency and performance in our proposed method, with 15.6% to 31.9% improvement in terms of effective traffic compared to other counterparts.

关键词： Resource management Optimization Vehicle dynamics Cloning Real-time systems multicast algorithms Autonomous vehicles Quality of service Heuristic algorithms Dynamic scheduling Age of information deep reinforcement learning multicastvehicular network

来源：评论

学校读者我要写书评

暂无评论

Secure MIMO Communication Relying on Movable Antennas

引用

IEEE TRANSACTIONS ON COMMUNICATIONS 2025年第4期73卷 2159-2175页

作者： Tang, Jun Pan, Cunhua Zhang, Yang Ren, Hong Wang, Kezhi Southeast Univ Natl Mobile Commun Res Lab Nanjing 210096 Peoples R China Brunel Univ London Dept Comp Sci Uxbridge UB8 3PH England

This paper considers a movable antenna (MA)-aided secure multiple-input multiple-output (MIMO) communication system consisting of a base station (BS), a legitimate information receiver (IR) and an eavesdropper (Eve), where the BS is equipped with MAs to enhance the system's physical layer security (PLS). Specifically, we aim to maximize the secrecy rate (SR) by jointly optimizing the transmit precoding (TPC) matrix, the artificial noise (AN) covariance matrix and the MAs' positions under the constraints of the maximum transmit power and the minimum spacing between MAs. To solve this non-convex problem with highly coupled optimization variables, the block coordinate descent (BCD) method is applied to alternately update the variables. Specifically, we first reformulate the SR into a tractable form, and derive the optimal TPC matrix and the AN covariance matrix with fixed MAs' positions by applying the Lagrangian multiplier method in semi-closed forms. Then, the majorization-minimization (MM) algorithm is employed to iteratively optimize each MA's position while keeping others fixed. We also extend this work to the more general multicast scenario. Finally, simulation results are provided to demonstrate the effectiveness of the proposed algorithms and the significant advantages of the MAs over conventional fixed position antennas (FPAs) in enhancing system's security.

关键词： Covariance matrices MIMO communication Antennas Security Optimization multicast algorithms Array signal processing Movable antenna (MA) physical layer security (PLS) artificial noise (AN) antenna position optimization

来源：评论

学校读者我要写书评

暂无评论

Enhancing 5G V2X URLLC Broadcast/multicast Services With FL-Based Wireless Resource Allocation

引用

IEEE TRANSACTIONS ON BROADCASTING 2025年

作者： Huang, Qian Yi, Xiaoyin Qi, Fei Liu, Lei Xie, Qingming Jiang, Qin Hu, Chunxia Chongqing Polytech Univ Elect Technol Artificial Intelligence & Big Data Coll Chongqing 401331 Peoples R China Chongqing Univ Posts & Telecommun Chongqing Key Lab Publ Big Data Secur Technol Chongqing 401420 Peoples R China Chongqing Univ Posts & Telecommun Chongqing Coll Mobile Commun Chongqing 401420 Peoples R China China Telecom Res Inst Dept Mobile Commun Technol Res Beijing 102209 Peoples R China Chongqing Polytech Univ Elect Technol Sch Smart Hlth Chongqing 401331 Peoples R China

This paper addresses the challenges of wireless resource allocation for 5G Ultra-reliable Low-latency Communication (URLLC) broadcast/multicast services in Vehicle-to-Everything (V2X) scenarios. It proposes three key algorithms: an iterative resource allocation approach that decomposes optimization into power and spectrum subproblems, a federated learning-based multicast resource allocation scheme that protects data privacy while enabling distributed training, and a cooperative multi-agent reinforcement learning solution that treats vehicles as intelligent nodes to jointly optimize system throughput, URLLC delivery rate, and multicast performance. Path loss models, mobility patterns, and interference scenarios are analyzed for both unicast and multicast transmissions. Simulation results demonstrate that the proposed algorithms achieve superior performance in terms of throughput, reliability, and latency compared to traditional and baseline approaches.

关键词： Resource management Ultra reliable low latency communication Vehicle-to-everything Reliability Mobility models multicast algorithms Unicast Throughput Bandwidth Training wireless resource allocation federated learning multi-agent reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

CALL SCHEDULING algorithms IN A multicast SWITCH

引用

IEEE TRANSACTIONS ON COMMUNICATIONS 1992年第3期40卷 625-635页

作者： KIM, CK LEE, TT POLYTECH INST NEW YORK DEPT ELECT ENGNBROOKLYNNY 11201

multicast switching is emerging as a new switching technology that can provide efficient transport in a broad-band network for video and other multipoint communication services. In this paper, we develop and analyze call scheduling algorithms for a multicast switch. In particular, we examine two general classes of scheduling algorithms: call packing algorithms and call splitting algorithms. Performance improvement by call packing algorithms examined in this paper is shown to be negligible. In contrast, call splitting algorithms can provide significantly lower blocking by reducing the level of output port contention. However, excessive call splitting could degrade performance because of the additional load introduced to the input ports. We present a simple call splitting algorithm called greedy splitting which achieves the performance approaching the best performance.

关键词： multicast algorithms Scheduling algorithm Switches Communication switching Switching circuits Algorithm design and analysis Intelligent networks Degradation Teleconferencing Hardware

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：