Article
Engineering, Electrical & Electronic
Chuan Sun, Xiuhua Li, Junhao Wen, Xiaofei Wang, Zhu Han, Victor C. M. Leung
Summary: To support the increasing services and applications, multi-tier computing architecture is used to distribute computing/caching/communication/networking capabilities between cloud servers and users. However, edge caching remains a serious issue due to heterogeneous content requests and high-cost direct hits. This paper proposes a recommendation-enabled edge caching framework that integrates recommender systems and edge caching to improve the resource utilization of edge servers. Simulation results show that the proposed framework outperforms existing algorithms in reducing average system cost.
IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS
(2023)
Article
Chemistry, Analytical
Yinghui Liu, Youyang Qu, Chenhao Xu, Zhicheng Hao, Bruce Gu
Summary: The rapid proliferation of edge computing devices has led to a growth in data, driving the development of machine learning technology, but privacy issues during data collection have raised concerns. To address these privacy concerns, synchronous federated learning method is proposed, but it suffers from inefficiencies and security vulnerabilities. Therefore, a novel method, Federated Learning with Asynchronous Convergence (FedAC) using a blockchain network, is proposed to address these issues.
Article
Computer Science, Theory & Methods
Carlo Mazzocca, Nicola Romandini, Rebecca Montanari, Paolo Bellavista
Summary: With the proliferation of IoT devices, edge cloud computing has become a promising paradigm to bring cloud capabilities closer to data sources. Federated Learning (FL) is emerging as a distributed ML approach that enables models to be trained on remote devices using their local data. However, traditional FL solutions still face challenges, and researchers are starting to propose leveraging blockchain technologies to address them.
FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE
(2024)
Article
Computer Science, Information Systems
Xinzhe Zheng, Yijie Zhang, Fan Yang, Fangmin Xu
Summary: This paper investigates the resource allocation problem with blockchain-based Mobile Edge Computing (MEC) system architecture. By applying a new consensus mechanism and optimizing algorithms, a more efficient and stable resource allocation policy is proposed.
Article
Computer Science, Hardware & Architecture
Rui Jin, Jia Hu, Geyong Min, Jed Mills
Summary: Federated Learning is a privacy-preserving distributed machine learning method, but it faces challenges like malicious clients and communication overhead. To address these challenges, a lightweight Blockchain-Empowered Federated Learning system is proposed, which integrates secure and efficient training scheme, consensus mechanism, and scalable blockchain architecture.
IEEE TRANSACTIONS ON COMPUTERS
(2023)
Article
Computer Science, Hardware & Architecture
Lei Peng, Zhixiang Yang, Shaoyong Guo, Xuesong Qiu, Wenjing Li, Peng Yu
Summary: The paper introduces a framework that combines blockchain and federated learning to address security and trust issues in FL on mobile edge networks. The framework includes a two-layered architecture with local and global model update chains, as well as a reputation-learning based incentive mechanism to make participant devices more trustworthy.
Article
Computer Science, Information Systems
Dinh C. Nguyen, Ming Ding, Quoc-Viet Pham, Pubudu N. Pathirana, Long Bao Le, Aruna Seneviratne, Jun Li, Dusit Niyato, H. Vincent Poor
Summary: The article discusses the concept of FLchain in MEC networks, focusing on privacy protection, security, cross-device collaboration, and resource allocation. FLchain integrates FL and blockchain technology, presenting a promising paradigm for intelligent MEC networks.
IEEE INTERNET OF THINGS JOURNAL
(2021)
Article
Engineering, Electrical & Electronic
Xinyu Ye, Meng Li, Pengbo Si, Ruizhe Yang, Zhuwei Wang, Yanhua Zhang
Summary: This paper introduces the application of blockchain technology in IoT vehicle's mobile edge computing system to solve the issues of data transmission privacy and authenticity, and designs an intelligent resource allocation framework to reduce energy consumption and improve the transactional throughput of the blockchain. The asynchronous advantage actor-critic approach is introduced to deal with the optimization problem. Simulation results show significant advantages of the proposed scheme over other comparison schemes.
IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY
(2023)
Article
Computer Science, Information Systems
Abla Smahi, Hui Li, Yong Yang, Xin Yang, Ping Lu, Yong Zhong, Caifu Liu
Summary: ICV generates a large amount of data in V2X environments, and FL can exploit this data effectively. Existing FL systems are vulnerable to attacks, but the proposed BV-ICVs framework uses blockchain and zkSNARKs verification to prevent unreliable model updates, increasing security and accuracy.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2023)
Article
Business
Hojjat Baghban, Amir Rezapour, Ching-Hsien Hsu, Sirapop Nuannimnoi, Ching-Yao Huang
Summary: This article discusses the critical role of edge computing in the IoT environment and proposes the concept of the edge federation. It introduces an intelligent reinforcement learning-based request service provisioning system and shows that it outperforms baseline approaches in terms of profit and response latency in the edge computing context.
IEEE TRANSACTIONS ON ENGINEERING MANAGEMENT
(2022)
Article
Computer Science, Information Systems
Binayak Kar, Widhi Yahya, Ying-Dar Lin, Asad Ali
Summary: The diverse computing paradigms of cloud, edge, and fog are needed to handle the vast amount of data generated by IoT devices. However, each paradigm has its own advantages and limitations. Cloud computing provides high computational power and storage capacity but suffers from high latency. Edge and fog computing offer lower latency but with limited capacity and coverage. A federation between these paradigms is required to meet the requirements of IoT devices and optimize traffic offloading.
IEEE COMMUNICATIONS SURVEYS AND TUTORIALS
(2023)
Article
Computer Science, Information Systems
Zhiyuan Wang, Hongli Xu, Jianchun Liu, Yang Xu, He Huang, Yangming Zhao
Summary: This paper proposes an efficient federated learning mechanism called FedCH to accelerate federated learning in heterogeneous edge computing. By constructing a special cluster topology and performing hierarchical aggregation, FedCH reduces completion time and network traffic compared to existing mechanisms.
IEEE TRANSACTIONS ON MOBILE COMPUTING
(2023)
Article
Computer Science, Information Systems
Yunlong Lu, Xiaohong Huang, Ke Zhang, Sabita Maharjan, Yan Zhang
Summary: This article introduces digital twin edge networks (DITENs) to bridge the gap between physical edge networks and digital systems, and proposes a blockchain-empowered federated learning scheme to enhance communication security and data privacy protection. Additionally, asynchronous aggregation and digital twin empowered reinforcement learning are used to improve the efficiency of the integrated scheme.
IEEE INTERNET OF THINGS JOURNAL
(2021)
Article
Chemistry, Multidisciplinary
Song Liu, Xiong Wang, Longshuo Hui, Weiguo Wu
Summary: This paper proposes a blockchain-based decentralized federated learning method called BD-FL to solve the issues of single-point server failure and lack of incentive mechanism in traditional federated learning. BD-FL combines blockchain and edge computing techniques and introduces an incentive mechanism to motivate local devices in participating actively. The experiment results demonstrate that BD-FL effectively reduces the model training time compared to baseline federated learning methods, and the R-PBFT algorithm improves the training efficiency of BD-FL.
APPLIED SCIENCES-BASEL
(2023)
Article
Computer Science, Information Systems
Laizhong Cui, Xiaoxin Su, Zhongxing Ming, Ziteng Chen, Shu Yang, Yipeng Zhou, Wei Xiao
Summary: This article introduces a system that combines IoT, edge computing, remote cloud, and blockchain. The authors propose a new algorithm called CREAT, which combines federated learning and blockchain technology to improve cache hit rate and data security.
IEEE INTERNET OF THINGS JOURNAL
(2022)
Article
Engineering, Electrical & Electronic
Bruno Clerckx, Yijie Mao, Eduard A. Jorswieck, Jinhong Yuan, David J. Love, Elza Erkip, Dusit Niyato
Summary: This Special Issue focuses on the theory, design, optimization, and applications of RS and RSMA in various network configurations. It begins with a tutorial paper written by a guest editor, which illustrates the basic principles and applications of RS and RSMA, followed by 17 technical papers.
IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS
(2023)
Article
Engineering, Electrical & Electronic
Bruno Clerckx, Yijie Mao, Eduard A. Jorswieck, Jinhong Yuan, David J. Love, Elza Erkip, Dusit Niyato
Summary: Rate-Splitting Multiple Access (RSMA) is evaluated for its importance in next generation communication systems as a powerful multiple access and interference management strategy. It offers numerous benefits and applications, addressing fundamental problems such as interference management and providing enhanced efficiency, universality, flexibility, robustness, and reliability. In 6G, RSMA can be applied to a wide range of scenarios and applications.
IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS
(2023)
Article
Engineering, Electrical & Electronic
Tzu-Hsuan Chou, Nicolo Michelusi, David J. Love, James V. Krogmeier
Summary: 6G operators may use mmWave and sub-THz bands to meet wireless access demand, but sub-THz communication faces new challenges due to wider bandwidths and harsher propagation conditions. This paper proposes a compressed training framework for estimating time-varying sub-THz MIMO-OFDM channels.
IEEE TRANSACTIONS ON COMMUNICATIONS
(2023)
Article
Computer Science, Information Systems
Dinh C. Nguyen, Ming Ding, Pubudu N. Pathirana, Aruna Seneviratne, Jun Li, H. Vincent Poor
Summary: This article proposes a novel cooperative task offloading and block mining scheme for blockchain-based MEC system, aiming to maximize system utility by jointly optimizing offloading decision, channel selection, transmit power allocation, and computational resource allocation. Simulation results demonstrate significant improvement of system utility compared to baseline approaches.
IEEE TRANSACTIONS ON MOBILE COMPUTING
(2023)
Article
Engineering, Electrical & Electronic
Zhang Liu, Minghui Liwang, Seyyedali Hosseinalipour, Huaiyu Dai, Zhibin Gao, Lianfen Huang
Summary: This paper investigates the challenges of scheduling tasks with a directed acyclic graph structure on dynamic vehicular cloud platforms and proposes a solution to minimize task completion time and ensure high success rate.
IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY
(2023)
Article
Computer Science, Information Systems
Minghui Liwang, Zhibin Gao, Seyyedali Hosseinalipour, Yuhan Su, Xianbin Wang, Huaiyu Dai
Summary: This article investigates the application of vehicular cloud-assisted task scheduling in an air-ground integrated vehicular network. By modeling tasks carried by unmanned aerial vehicles and resources of vehicular clouds as graph structures, the authors consider the scenario where resource-limited UAVs offload computation-intensive tasks to resource-abundant vehicles for processing. They formulate an optimization problem to jointly optimize the mapping between task components and vehicles, transmission powers of UAVs, and address the trade-off between completion time of tasks, energy consumption of UAVs, and data exchange cost among vehicles. The authors propose a decoupling approach for task scheduling by segregating template searching from transmission power allocation, which is shown to outperform baseline methods in extensive simulations.
IEEE TRANSACTIONS ON SERVICES COMPUTING
(2023)
Article
Computer Science, Information Systems
Su Wang, Seyyedali Hosseinalipour, Maria Gorlatova, Christopher G. Brinton, Mung Chiang
Summary: In this study, we train machine learning models on geo-distributed, resource-constrained clusters of devices through unmanned aerial vehicle (UAV) swarms. We address the challenges posed by varying data heterogeneity and computational resource inadequacy among device clusters by introducing stratified UAV swarms, hierarchical nested personalized federated learning (HN-PFL), cooperative UAV resource pooling, and model/concept drift. Our methodology considers both micro and macro system design, with a focus on network-aware HN-PFL and swarm trajectory and learning duration design tackled via deep reinforcement learning. Simulations demonstrate the effectiveness of our approach in terms of ML performance, resource savings, and swarm trajectory efficiency.
IEEE TRANSACTIONS ON NETWORK AND SERVICE MANAGEMENT
(2023)
Article
Computer Science, Hardware & Architecture
Seyyedali Hosseinalipour, Su Wang, Nicolo Michelusi, Vaneet Aggarwal, Christopher G. Brinton, David J. Love, Mung Chiang
Summary: In this paper, we propose parallel successive learning (PSL) to expand the architecture of federated learning in terms of network, heterogeneity, and proximity. PSL considers decentralized cooperation among devices, heterogeneous learning and data environments, and devices with different capabilities. We also analyze the concepts of cold vs. warmed up models and propose a network-aware dynamic model tracking method to optimize the tradeoff between model learning and resource efficiency. Our numerical results reveal new insights on the interdependencies between idle times, model/concept drift, and D2D cooperation configuration.
IEEE-ACM TRANSACTIONS ON NETWORKING
(2023)
Article
Computer Science, Hardware & Architecture
Payam Abdisarabshali, Minghui Liwang, Amir Rajabzadeh, Mahmood Ahmadi, Seyyedali Hosseinalipour
Summary: Vehicular cloud is a promising technology for processing computation-intensive applications on smart vehicles. This work addresses the challenges of insufficient computing resources and dynamic resource availability caused by vehicle mobility in implementing vehicular clouds. A general reliability metric and a redundancy-based processing framework are introduced to improve the reliability of CI-App processing. A mathematical framework called event stochastic algebra is developed to analyze the reliability of the proposed methodology. Simulation results demonstrate the accuracy and efficiency of the proposed methodology in CI-App processing.
IEEE-ACM TRANSACTIONS ON NETWORKING
(2023)
Article
Computer Science, Hardware & Architecture
Bhargav Ganguly, Seyyedali Hosseinalipour, Kwang Taik Kim, Christopher G. Brinton, Vaneet Aggarwal, David J. Love, Mung Chiang
Summary: We propose a cooperative edge-assisted dynamic federated learning (CE-FL) approach. CE-FL introduces a distributed machine learning (ML) architecture, where data collection is carried out at the end devices, while the model training is conducted cooperatively at the end devices and the edge servers, enabled via data offloading from the end devices to the edge servers through base stations. CE-FL also introduces a floating aggregation point, where the local models generated at the devices and the servers are aggregated at an edge server, which varies from one model training round to another to cope with the network evolution in terms of data distribution and users' mobility.
IEEE-ACM TRANSACTIONS ON NETWORKING
(2023)
Article
Computer Science, Theory & Methods
Dinh C Nguyen, Quoc-Viet Pham, Pubudu N. Pathirana, Ming Ding, Aruna Seneviratne, Zihuai Lin, Octavia Dobre, Won-Joo Hwang
Summary: Recent advances in communication technologies and the Internet-of-Medical-Things (IOMT) have enabled the use of artificial intelligence (AI) in smart healthcare. Federated Learning (FL), as a distributed collaborative AI paradigm, is particularly attractive for smart healthcare due to its ability to train AI models without sharing raw data. This survey provides a comprehensive overview of the recent advances in FL, its motivations, requirements, and applications in key healthcare domains.
ACM COMPUTING SURVEYS
(2023)
Article
Computer Science, Information Systems
Thang Ngo, Pubudu N. N. Pathirana, Malcolm K. K. Horne, Louise A. A. Corben, Ian H. H. Harding, David J. J. Szmulewicz
Summary: Cerebellar ataxia is a movement disorder caused by injury or disease to the cerebellum. Diagnosis and assessment of ataxia are challenging due to the reliance on clinical experience and subjectivity. Recent advancements in neuroimaging, sensor-based approaches, and machine learning techniques have shown promise in addressing these challenges. This paper provides an overview of the clinical challenges and outlines possible machine learning approaches, while discussing limitations and potential for future research.
Article
Engineering, Electrical & Electronic
Jing Guo, Raghu G. Raj, David J. Love, Christopher G. Brinton
Summary: This paper focuses on the signal classification problem in wireless sensor networks. It proposes a multi-sensor online kernel scalar quantization learning strategy to maximize classification performance and improve network resource efficiency through sparse sensor selection using a marginalized weighted kernel approach.
IEEE TRANSACTIONS ON SIGNAL PROCESSING
(2022)