Profit-Aware Cooperative Offloading in UAV-Enabled MEC Systems Using Lightweight Deep Reinforcement Learning DOI
Zheyi Chen, Junjie Zhang, Xianghan Zheng

et al.

IEEE Internet of Things Journal, Journal Year: 2023, Volume and Issue: 11(12), P. 21325 - 21336

Published: Nov. 10, 2023

In Mobile Edge Computing (MEC) systems, Unmanned Aerial Vehicles (UAVs) facilitate Service Providers (ESPs) offering flexible resource provisioning with broader communication coverage and thus improving the Quality-of-Service (QoS). However, dynamic system states various traffic patterns seriously hinder efficient cooperation among UAVs. Existing solutions commonly rely on prior knowledge or complex neural network models, lacking adaptability causing excessive overheads. To address these critical challenges, we propose DisOff, a novel profit-aware cooperative offloading framework in UAV-enabled MEC lightweight Deep Reinforcement Learning (DRL). First, design an improved DRL twin critic-networks delay mechanism, which solves Q-value overestimation high variance approximates optimal UAV allocation. Next, develop new multi-teacher distillation mechanism for proposed model, where policies of multiple UAVs are integrated into one agent, compressing model size while maintaining superior performance. Using real-world datasets user traffic, extensive experiments conducted to validate effectiveness DisOff. Compared benchmark methods, DisOff enhances ESP profits reducing training costs.

Language: Английский

A data-driven approach for intrusion and anomaly detection using automated machine learning for the Internet of Things DOI

Hao Xu,

Zihan Sun,

Yuan Cao

et al.

Soft Computing, Journal Year: 2023, Volume and Issue: 27(19), P. 14469 - 14481

Published: July 29, 2023

Language: Английский

Citations

148
Networking Architectures and Protocols for IoT Applications in Smart Cities: Recent Developments and Perspectives DOI Open Access
Dimitris Kanellopoulos,

Varun Kumar Sharma,

Theodor Panagiotakopoulos

et al.

Electronics, Journal Year: 2023, Volume and Issue: 12(11), P. 2490 - 2490

Published: May 31, 2023

Numerous municipalities employ the smart city model in large cities to improve quality of life their residents, utilize local resources efficiently, and save operating expenses. This incorporates many heterogeneous technologies such as Cyber-Physical Systems (CPS), Wireless Sensor Networks (WSNs), Cloud Computing (ClCom). However, effective networking communication protocols are required provide essential harmonization control system mechanisms achieve these crucial goals. The requirements characteristics applications (SCAs) identified this study, well that can be utilized serve diverse data traffic flows between dissimilar mechanisms. Additionally, we show examples designs a few systems, transport, building, home, grid, water, pipeline monitoring, systems.

Language: Английский

Citations

47
Adaptive Segmentation Enhanced Asynchronous Federated Learning for Sustainable Intelligent Transportation Systems DOI
Xiaokang Zhou, Wei Liang, Akira Kawai

et al.

IEEE Transactions on Intelligent Transportation Systems, Journal Year: 2024, Volume and Issue: 25(7), P. 6658 - 6666

Published: Feb. 19, 2024

The proliferation of advanced embedded and communication technologies has facilitated the possibility modern Intelligent Transportation System (ITS). hierarchical nature such large-scale distributed systems brings obvious challenges in creating a scalable sustainable computing environment, hence development application edge intelligence become critical. Federated learning (FL), as an emerging machine paradigm, aims to offer secure knowledge sharing effective across multiple devices. However, conventional FL may fall into trouble when facing network-agnostic with fast moving devices changing network attributes. In this study, we propose Adaptive Segmentation enhanced Asynchronous Learning (AS-AFL) model, aiming improve efficiency reliability ITS via decentralized fashion. Specifically, meta-learning based adaptive segmentation scheme is designed automatically separate client nodes (e.g., vehicles) groups according their homogeneous An integrated aggregation mechanism then developed realize horizontal among group similar so-called intra-group synchronous aggregation, while allowing vertical different inter-group asynchronous aggregation. Experiment evaluation results on open-source dataset demonstrate outstanding performance our proposed compared several schemes scenario.

Language: Английский

Citations

27
CMAF-IIoT: Chaotic map-based authentication framework for Industrial Internet of Things DOI
Muhammad Tanveer, Akhtar Badshah, Abd Ullah Khan

et al.

Internet of Things, Journal Year: 2023, Volume and Issue: 23, P. 100902 - 100902

Published: Aug. 10, 2023

Language: Английский

Citations

39
Safe-NORA: Safe Reinforcement Learning-based Mobile Network Resource Allocation for Diverse User Demands DOI Open Access
Wenzhen Huang, Tong Li, Yuting Cao

et al.

Published: Oct. 21, 2023

As mobile communication technologies advance, networks become increasingly complex, and user requirements diverse. To satisfy the diverse demands of users while improving overall performance network system, limited wireless resources should be efficiently dynamically allocated to them based on magnitude their relative location base stations. We separated problem into four constrained subproblems, which we then solved using a safe reinforcement learning method. In addition, design reward mechanism encourage agent cooperation in distributed training environments. test our methodology simulated scenario with thousands hundreds According experimental findings, method guarantees that over 95% are satisfied also maximizing system throughput.

Language: Английский

Citations

34
Small wind turbines and their potential for internet of things applications DOI Creative Commons
Hao Wang,

Bendong Xiong,

Zutao Zhang

et al.

iScience, Journal Year: 2023, Volume and Issue: 26(9), P. 107674 - 107674

Published: Aug. 18, 2023

Wind energy is crucial for meeting climate and sustainability targets. Small wind turbines (SWTs) have gained significant attention due to their size adaptability. These potential Internet of Things (IoT) applications, particularly in powering large areas low-power devices. This review examines SWTs IoT providing an extensive overview development, including rectifiers, power generation mechanisms, applications. The paper summarizes compares different types explores recent advancements representative work, discusses applicable generator systems such as electromagnetic, piezoelectric, triboelectric nanogenerators. In addition, it thoroughly reviews the latest research on application scenarios, transportation, urban environments, intelligent agriculture, self-powered sensing. Lastly, identifies future directions emphasizes interdisciplinary technologies driving SWT development.

Language: Английский

Citations

31
A Survey on Network Slicing Security: Attacks, Challenges, Solutions and Research Directions DOI Creative Commons
Chamitha de Alwis, Pawani Porambage, Kapal Dev

et al.

IEEE Communications Surveys & Tutorials, Journal Year: 2023, Volume and Issue: 26(1), P. 534 - 570

Published: Sept. 6, 2023

The dawn of softwarized networks enables Network Slicing (NS) as an important technology towards allocating end-to-end logical to facilitate diverse requirements emerging applications in fifth-generation (5G) mobile networks. However, the emergence NS also exposes novel security and privacy challenges, primarily related aspects such life-cycle security, inter-slice intra-slice slice broker zero-touch network management blockchain security. Hence, enhancing privacy, trust has become a key research area toward realizing true capabilities 5G. This paper presents comprehensive up-to-date survey on articulates taxonomy for laying structure survey. Accordingly, attack scenarios specific NS-enabled Furthermore, explores threats, issues while elaborating solutions available literature. In addition, aspects, along with possible solutions, are explained. highlights future directions privacy. It is envisaged that this will concentrate existing work, highlight gaps shed light research, development, standardization work realize secure 5G beyond communication

Language: Английский

Citations

31
Reliable Resource Allocation and Management for IoT Transportation Using Fog Computing DOI Open Access

Haseeb Ullah Atiq,

Zulfiqar Ahmad,

Sardar Khaliq uz Zaman

et al.

Electronics, Journal Year: 2023, Volume and Issue: 12(6), P. 1452 - 1452

Published: March 19, 2023

Resource allocation in smart settings, more specifically Internet of Things (IoT) transportation, is challenging due to the complexity and dynamic nature fog computing. The demands users may alter over time, necessitating trustworthy resource administration. Effective management systems must be designed accommodate changing user needs. Fog devices don’t just run fog-specific software. link failures could brought on by absence centralised administration, device autonomy, wireless communication environment. Resources allocated managed effectively because majority are battery-powered. Latency-aware IoT applications, such as intelligent healthcare, emergency response, now pervasive a result enormous growth ubiquitous These services generate large amount data, which requires edge processing. flexibility on-demand for cloud can successfully manage these applications. It’s not always advisable applications exclusively cloud, especially latency-sensitive Thus, computing has emerged bridge between it supports. This typically how sensors connected. neighbouring control storage intermediary computation. In order improve environment reliability IoT-based systems, this paper suggests strategy. When assigning resources, latency energy efficiency taken into account. Users prioritise cost-effectiveness speed fog. Simulation was performed iFogSim2 simulation tool, performance compared with one existing state-of-the-art A comparison results shows that proposed strategy reduced 10.3% consumption 21.85% when

Language: Английский

Citations

29
Edu-Metaverse: concept, architecture, and applications DOI
Ting Wu, Fei Hao

Interactive Learning Environments, Journal Year: 2023, Volume and Issue: 32(8), P. 4352 - 4379

Published: April 17, 2023

As an important form of education in the future, Edu-Metaverse cannot only innovate existing teaching mode, provide diversified resources and environments, realize intelligent evaluation certification methods, etc., but also real integration people's physical world virtual world. Therefore, how to deepen concept metaverse into field creating innovative model has become challenge this research. To end, paper develops a comprehensive ecosystem architecture which provides effective implementation reference for development Edu-Metaverse. In addition, mode is constructed, feasibility its breakthrough significance offline online are demonstrated through questionnaires. Through case study teaching, scheme ways verified. The research shows that can not broaden educational content, reduce cost improve quality efficiency education, it potential risks.

Language: Английский

Citations

26
Traffic-Aware Reliable Scheduling in TSCH Networks for Industry 4.0: A Systematic Mapping Review DOI
Abdeldjalil Tabouche, Badis Djamaa, Mustapha Réda Senouci

et al.

IEEE Communications Surveys & Tutorials, Journal Year: 2023, Volume and Issue: 25(4), P. 2834 - 2861

Published: Jan. 1, 2023

Recently, mission-critical Industrial Internet of Things (IIoT) applications such as system automation, predictive maintenance, and anomaly detection have come into the spotlight Industry 4.0 thanks to promised benefits. The IEEE 802.15.4 Time-Slotted Channel Hopping (TSCH) mode, along with IPv6 over TSCH (6TiSCH) initiative, are two key standards accommodate diverse traffic patterns, reliability, latency, power efficiency needs IIoT applications. To manage allocation communication resources in networks, a Scheduling Function (SF) is implemented. Even though scheduling has been subject numerous reviews, potential taking traffic-awareness account not fully investigated. Motivated by these facts, we classify analyze, this systematic mapping review, prominent SFs dealing networks published between 2012 2022. As result, provide multi-dimensional map identify current trends traffic-aware help assess how far given proposal supported or contradicted empirical evidence field. Consequently, discuss some open challenges that require community attention point out future research directions regarding design, implementation, evaluation new SFs.

Language: Английский

Citations

23