检索结果-内蒙古大学图书馆

Toward Mixture-of-Experts Enabled Trustworthy Semantic Communication for 6G Networks

IEEE Network 2024年

作者： He, Jiayi Luo, Xiaofeng Kang, Jiawen Du, Hongyang Xiong, Zehui Chen, Ci Niyato, Dusit Shen, Xuemin Guangdong University of Technology School of Automation Guangzhou510006 China Nanyang Technological University College of Computing and Data Science Singapore639798 Singapore Singapore University of Technology and Design Pillar of Information Systems Technology and Design Singapore487372 Singapore University of Waterloo Department of Electrical and Computer Engineering WaterlooONN2L 3G1 Canada

Semantic Communication (SemCom) plays a pivotal role in 6G networks, offering a viable solution for future efficient communication. Through nonlinear mapping of semantic representations, Deep Learning (DL)-based semantic communications further enhance communication efficiency. However, the vulnerability of DL models to security threats, such as adversarial attacks, poses significant challenges for practical applications of SemCom systems. These vulnerabilities allow attackers to tamper with messages and eavesdrop on private information, especially in wireless communication scenarios. Although existing defenses attempt to address specific threats, they often fail to simultaneously handle multiple heterogeneous attacks. To overcome this limitation, we propose a novel Mixture-of-Experts (MoE)-based SemCom system, comprising a gating network and multiple experts, each specializing in different security challenges. The gating network adaptively selects suitable experts to counter heterogeneous attacks based on user-defined security requirements. Multiple experts collaborate to accomplish semantic communication tasks while meeting the security requirements of users. A case study in vehicular networks demonstrates the efficacy of the MoE-based SemCom system. Simulation results show that the proposed MoE-based SemCom system effectively mitigates concurrent heterogeneous attacks, with minimal impact on downstream task accuracy. © 1986-2012 IEEE.

关键词： Vehicular ad hoc networks

来源：评论

学校读者我要写书评

暂无评论

How to design a R-D Model for UAV Video Transmission?

How to Design a R-D Model for UAV Video Transmission?

引用

International Conference on Wireless Communications and Signal Processing (WCSP)

作者： Zongyang Yu Peng Yang Xianbin Cao Dezhi Zheng Tony Q. S. Quek Dapeng Oliver Wu School of Electronic Information Engineering Beihang University Beijing China Department of Mathematics and Theories Peng Cheng Laboratory Shenzhen China Advanced Research Institute of Multidisciplinary Sciences Beijing Institute of Technology Beijing China Information Systems Technology and Design Singapore University of Technology and Design Singapore Department of Computer Science City University of Hong Kong Kowloon Hong Kong

Unmanned aerial vehicle (UAV) video transmission has been extensively applied in many crucial fields. However, the problem of designing a rate-distortion (R-D) model for UAV video transmission, which is essential for theoretical analysis of video coding and optimization of video transmission quality, is under-studied. The designed R-D model is desired to be simple, accurate, and generic owing to the limited capability of the UAV. Observing the key role of transformed residuals in the R-D model, this paper elaborately discusses the modeling of the transformed residual distribution and proposes an estimation algorithm to estimate the statistical parameter of the distribution. Specifically, considering the stringent requirements for low complexity and high generalization capability, we first design a linear parameter estimator by mining and analyzing UAV video statistical characteristics. Further, we develop a bias update scheme to improve the accuracy of the estimator. Test results on multiple real video sequences taken by UAVs show that the proposed estimation algorithm is more accurate than benchmarks.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Generative AI Agents with Large Language Model for Satellite Networks via a Mixture of Experts Transmission

arXiv

引用

arXiv 2024年

作者： Zhang, Ruichen Du, Hongyang Liu, Yinqiu Niyato, Dusit Kang, Jiawen Xiong, Zehui Jamalipour, Abbas Kim, Dong In The College of Computing and Data Science Nanyang Technological University Singapore The School of Automation Guangdong University of Technology China The Pillar of Information Systems Technology and Design Singapore University of Technology and Design Singapore The University of Sydney SydneyNSW2006 Australia The Department of Electrical and Computer Engineering Sungkyunkwan University Suwon16419 Korea Republic of

In response to the needs of 6G global communications, satellite communication networks have emerged as a key solution. However, the large-scale development of satellite communication networks is constrained by the complex system models, whose modeling is challenging for massive users. Moreover, transmission interference between satellites and users seriously affects communication performance. To solve these problems, this paper develops generative artificial intelligence (AI) agents for model formulation and then applies a mixture of experts (MoE) approach to design transmission strategies. Specifically, we leverage large language models (LLMs) to build an interactive modeling paradigm and utilize retrieval-augmented generation (RAG) to extract satellite expert knowledge that supports mathematical modeling. Afterward, by integrating the expertise of multiple specialized components, we propose an MoE-proximal policy optimization (PPO) approach to solve the formulated problem. Each expert can optimize the optimization variables at which it excels through specialized training through its own network and then aggregates them through the gating network to perform joint optimization. The simulation results validate the accuracy and effectiveness of employing a generative agent for problem formulation. Furthermore, the superiority of the proposed MoE-ppo approach over other benchmarks is confirmed in solving the formulated problem. The adaptability of MoE-PPO to various customized modeling problems has also been demonstrated. © 2024, CC BY.

关键词： Satellite communication systems

来源：评论

学校读者我要写书评

暂无评论

Optimal AoI-based Block Propagation and Incentive Mechanism for Blockchain Networks in Web 3.0

引用

IEEE Transactions on Cognitive Communications and Networking 2024年

作者： Wen, Jinbo Kang, Jiawen Xiong, Zehui Du, Hongyang Yang, Zhaohui Niyato, Dusit Shen, Meng Jiao, Yutao Zhang, Yang Nanjing University of Aeronautics and Astronautics College of Computer Science and Technology China Guangdong University of Technology School of Automation China Singapore University of Technology and Design Pillar of Information Systems Technology and Design Singapore Hong Kong University Department of Electrical and Electronic Engineering Hong Kong Zhejiang University College of Information Science and Electronic Engineering China Nanyang Technological University College of Computing and Data Science Singapore Beijing Institute of Technology School of Cyberspace Science and Technology China Army Engineering University of PLA College of Communications Engineering China

Web 3.0 is regarded as a revolutionary paradigm that enables users to securely manage data without a centralized authority. Blockchains, which enable data to be managed in a decentralized and transparent manner, are key technologies for achieving Web 3.0 goals. However, blockchain-based Web 3.0 is still in its infancy, such as ensuring block freshness and optimizing block propagation to improve blockchain performance. In this paper, we develop a freshness-aware block propagation optimization framework for Web 3.0. We first propose a novel metric called Age of Block information (AoBI) based on the concept of age of information to quantify block freshness. AoBI measures the time elapsed from the freshest transaction generation to the completion of block consensus. To make block propagation optimization tractable, we classify miners into five different states and propose a block propagation model for public blockchains inspired by epidemic models. Moreover, considering that the miners are bounded rational, we propose an incentive mechanism based on the evolutionary game for block propagation to improve block propagation efficiency. Numerical results demonstrate that compared with other block propagation mechanisms in public blockchains, the proposed scheme has a higher block forwarding probability, which improves block propagation efficiency and decreases the minimum value of average AoBI. © 2015 IEEE.

关键词： age of information block propagation evolutionary game Web 3.0 wireless networks

来源：评论

学校读者我要写书评

暂无评论

A K -Band Four-Beam Transmitter with Decoupled Phase/Gain Control and Enhanced Power Back off Efficiency for SATCOM

引用

IEEE Transactions on Microwave Theory and Techniques 2024年第11期72卷 6443-6459页

作者： Gao, Huiyan Li, Nayu Xie, Xinhong Wang, Shaogang Lu, Hang Kuan, Yen-Cheng Yu, Xiaopeng Song, Chunyi Jane Gu, Qun Xu, Zhiwei Institute of Marine Electronics and Intelligent Systems Ocean College Zhejiang University Zhoushan316021 China Engineering Research Center of Oceanic Sensing Technology and Equipment Ministry of Education Zhoushan316021 China Donghai Laboratory Zhoushan316021 China National Yang Ming Chiao Tung University International College of Semiconductor Technology Hsinchu30010 Taiwan Institute of VLSI Design and Research College of Information Science and Electronic Engineering Zhejiang University Hangzhou310058 China University of California at Davis Department of Electrical and Computer Engineering DavisCA95616 United States

This article presents a 17.7-20.2 GHz eight-element four-beam RF-beamforming transmitter in 65-nm CMOS for satellite communication (SATCOM). The transmitter utilizes an analog scheme in the variable-gain amplifier (VGA) to achieve dB-in-linear gain control with a high dynamic range (DR), together with a current combining crossbar for a compact footprint. The vector-modulation-based (VM-based) phase shifter (PS) adopts the cancellation-based Gm units for low switching variation. Furthermore, eight PSs of each beam are split into two groups, with each group sharing the same tunable I/Q generator for low power consumption. Considering the increased peak-to-average power ratio (PAPR) caused by high-order modulation schemes and multibeam operation, the transmitter incorporates compact transformer-based Doherty power amplifiers (DPAs) with adaptive bias control to improve the power efficiency at back-off. The chip achieves a 31-dB gain-tuning range with a 0.283-dB step and a 360° phase-shifting range with a calibrated 7-bit resolution. The measured root-mean-square (RMS) phase/gain errors induced by the VGA/PS are below 2.05°/0.147 dB, respectively, which decouples phase/gain control. The transmitter achieves a maximum output 1-dB gain compression point (OP1dB) of 17.9 dBm and a maximum saturated power (Psat) of 18.9 dBm. The output element exhibits 29.3% drain efficiency (DE) at 6-dB power back off (PBO) from Psat , which is 1.56/3.12 times over the normalized class-B/class-A implementations, demonstrating Doherty PBO efficiency enhancement. The chip occupies an area of 8.59 × 4.17 mm2 and consumes 398 mW per element at OP1dB. Additionally, a 16-element linear array has been developed to showcase the capabilities of the multibeam operation and gain tapering. © 1963-2012 IEEE.

关键词： Efficiency

来源：评论

学校读者我要写书评

暂无评论

Generating adjacency-constrained subgoals in hierarchical reinforcement learning

arXiv

引用

arXiv 2020年

作者： Zhang, Tianren Guo, Shangqi Tan, Tian Hu, Xiaolin Chen, Feng Department of Automation Tsinghua University China Department of Civil and Environmental Engineering Stanford University United States Department of Computer Science and Technology Tsinghua University China Beijing National Research Center for Information Science and Technology China State Key Laboratory of Intelligent Technology and Systems Beijing Innovation Center for Future Chip China LSBDPA Beijing Key Laboratory China

Goal-conditioned hierarchical reinforcement learning (HRL) is a promising approach for scaling up reinforcement learning (RL) techniques. However, it often suffers from training inefficiency as the action space of the high-level, i.e., the goal space, is often large. Searching in a large goal space poses difficulties for both high-level subgoal generation and low-level policy learning. In this paper, we show that this problem can be effectively alleviated by restricting the high-level action space from the whole goal space to a k-step adjacent region of the current state using an adjacency constraint. We theoretically prove that the proposed adjacency constraint preserves the optimal hierarchical policy in deterministic MDPs, and show that this constraint can be practically implemented by training an adjacency network that can discriminate between adjacent and non-adjacent subgoals. Experimental results on discrete and continuous control tasks show that incorporating the adjacency constraint improves the performance of state-of-the-art HRL approaches in both deterministic and stochastic environments.1 Copyright © 2020, The Authors. All rights reserved.

关键词： Stochastic systems

来源：评论

学校读者我要写书评

暂无评论

Grasping simulation of an underactuated finger mechanism for larm hand

引用

International Journal of Modelling and Simulation 2010年第1期30卷 87-97页

作者： Yao, S. Wu, L. Ceccarelli, M. Carbone, G. Lu, Z. School of Automation Science and Electrical Engineering Beihang University Beijing China College of Information Engineering Central University for Nationalities Beijing China State Key Laboratory of Intelligent Technology and Systems Department of Computer Science and Technology Tsinghua University Beijing China Beijing University of Aeronautics and Astronautics China and University of Cassino Italy LARM: Laboratory of Robotics and Mechatronics Cassino United States

In this paper, the grasp operation of an underactuated finger mechanism is simulated with the aim to characterize and analyze its performance. The design of the proposed mechanism is based on spring systems that give suitable underactuation characteristics. A CAD model is proposed and then it is used in a procedure for numerical simulation. Simulation results show suitable capability for grasping a wide range of objects in size and shape by operating the mechanism inside the finger body. The achieved grasp has been simulated as a human-like functionality with a grasp contact in each phalanx.

关键词： computer aided design

来源：评论

学校读者我要写书评

暂无评论

Robust Generalization Against Photon-Limited Corruptions via Worst-Case Sharpness Minimization

Robust Generalization Against Photon-Limited Corruptions via...

引用

Conference on computer Vision and Pattern Recognition (CVPR)

作者： Zhuo Huang Miaoxi Zhu Xiaobo Xia Li Shen Jun Yu Chen Gong Bo Han Bo Du Tongliang Liu Sydney AI Centre The University of Sydney National Engineering Research Center for Multimedia Software Institute of Artificial Intelligence School of Computer Science Hubei Key Laboratory of Multimedia and Network Communication Engineering Wuhan University JD Explore Academy Department of Automation University of Science and Technology of China Key Laboratory of Intelligent Perception and Systems for High-Dimensional Information of Ministry of Education School of Computer Science and Engineering Nanjing University of Science and Technology Department of Computer Science Hong Kong Baptist University

Robust generalization aims to tackle the most challenging data distributions which are rare in the training set and contain severe noises, i.e., photon-limited corruptions. Common solutions such as distributionally robust optimization (DRO) focus on the worst-case empirical risk to ensure low training error on the uncommon noisy distributions. However, due to the over-parameterized model being optimized on scarce worst-case data, DRO fails to produce a smooth loss landscape, thus struggling on generalizing well to the test set. Therefore, instead of focusing on the worst-case risk minimization, we propose SharpDRO by penalizing the sharpness of the worst-case distribution, which measures the loss changes around the neighbor of learning parameters. Through worst-case sharpness minimization, the proposed method successfully produces a flat loss curve on the corrupted distributions, thus achieving robust generalization. Moreover, by considering whether the distribution annotation is available, we apply SharpDRO to two problem settings and design a worst-case selection process for robust generalization. Theoretically, we show that SharpDRO has a great convergence guarantee. Experimentally, we simulate photon-limited corruptions using CIFAR10/100 and ImageNet30 datasets and show that SharpDRO exhibits a strong generalization ability against severe corruptions and exceeds well-known baseline methods with large performance gains.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Mixture of Experts for Intelligent Networks: A Large Language Model-enabled Approach

Mixture of Experts for Intelligent Networks: A Large Languag...

引用

International Wireless Communications and Mobile Computing Conference, IWCMC

作者： Hongyang Du Guangyuan Liu Yijing Lin Dusit Niyato Jiawen Kang Zehui Xiong Dong In Kim School of Computer Science and Engineering Nanyang Technological University Singapore Singapore State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunications Beijing China School of Automation Guangdong University of Technology and Key Laboratory of Intelligent Information Processing and System Integration of IoT Ministry of Education Guangzhou China Guangdong-HongKong-Macao Joint Laboratory for Smart Discrete Manufacturing Guangzhou China Pillar of Information Systems Technology and Design Singapore University of Technology and Design Singapore Singapore Department of Electrical and Computer Engineering Sungkyunkwan University Suwon South Korea

ISBN: (数字)9798350361261

ISBN: (纸本)9798350361278

Optimizing various wireless user tasks poses a significant challenge for networking systems because of the expanding range of user requirements. Despite advancements in Deep Reinforcement Learning (DRL), the need for customized optimization tasks for individual users complicates developing and applying numerous DRL models, leading to substantial computation resource and energy consumption and can lead to inconsistent outcomes. To address this issue, we propose a novel approach utilizing a Mixture of Experts (MoE) framework, augmented with Large Language Models (LLMs), to analyze user objectives and constraints effectively, select specialized DRL experts, and weigh each decision from the participating experts. Specifically, we develop a gate network to oversee the expert models, allowing a collective of experts to tackle a wide array of new tasks. Furthermore, we innovatively substitute the traditional gate network with an LLM, leveraging its advanced reasoning capabilities to manage expert model selection for joint decisions. Our proposed method reduces the need to train new DRL models for each unique optimization problem, decreasing energy consumption and AI model implementation costs. The LLMenabled MoE approach is validated through a general maze navigation task and a specific network service provider utility maximization task, demonstrating its effectiveness and practical applicability in optimizing complex networking systems.

关键词： Wireless communication Energy consumption Navigation Computational modeling Logic gates Probability Throughput

来源：评论

学校读者我要写书评

暂无评论

Exploring Collaborative Distributed Diffusion-Based AI-Generated Content (AIGC) in Wireless Networks

arXiv

引用

arXiv 2023年

作者： Du, Hongyang Zhang, Ruichen Niyato, Dusit Kang, Jiawen Xiong, Zehui Kim, Dong In Shen, Xuemin Poor, H. Vincent The School of Computer Science and Engineering The Energy Research Institute @ NTU Interdisciplinary Graduate Program Nanyang Technological University Singapore The School of Computer Science and Engineering Nanyang Technological University Singapore The School of Computer and Information Technology Beijing Jiaotong University Beijing100044 China The School of Automation Guangdong University of Technology China The Pillar of Information Systems Technology and Design Singapore University of Technology and Design Singapore The Department of Electrical and Computer Engineering Sungkyunkwan University Suwon16419 Korea Republic of The Department of Electrical and Computer Engineering University of Waterloo Canada The Department of Electrical and Computer Engineering Princeton University PrincetonNJ08544 United States

Driven by advances in generative artificial intelligence (AI) techniques and algorithms, the widespread adoption of AI-generated content (AIGC) has emerged, allowing for the generation of diverse and high-quality content. Especially, the diffusion model-based AIGC technique has been widely used to generate content in a variety of modalities. However, the real-world implementation of AIGC models, particularly on resource-constrained devices such as mobile phones, introduces significant challenges related to energy consumption and privacy concerns. To further promote the realization of ubiquitous AIGC services, we propose a novel collaborative distributed diffusion-based AIGC framework. By capitalizing on collaboration among devices in wireless networks, the proposed framework facilitates the efficient execution of AIGC tasks, optimizing edge computation resource utilization. Furthermore, we examine the practical implementation of the denoising steps on mobile phones, the impact of the proposed approach on the wireless network-aided AIGC landscape, and the future opportunities associated with its real-world integration. The contributions of this paper not only offer a promising solution to the existing limitations of AIGC services but also pave the way for future research in device collaboration, resource optimization, and the seamless delivery of AIGC services across various devices. Our code is available at https://***/HongyangDu/DistributedDiffusion. © 2023, CC BY.

关键词： Wireless networks

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：