Scheduling for Minimizing the Age of Information in Multisensor Multiserver Industrial Internet of Things Systems DOI
Xin Xie, Heng Wang, Xiaojiang Liu

et al.

IEEE Transactions on Industrial Informatics, Journal Year: 2023, Volume and Issue: 20(1), P. 573 - 582

Published: April 20, 2023

Real-time data delivery is significant for the Industrial Internet of Things (IIoT). Age information (AoI), a popular real-time metric, usually used to measure freshness IIoT systems. If most recently received by destination at time $t$ was generated notation="LaTeX">$t_{1}$ , then AoI notation="LaTeX">$t-t_{1}$ . In this paper, we consider multi-sensor multi-server system and develop scheduling algorithms minimize average AoI. The challenge lies in strong coupling between link scheduling, server selection, service preemption. To address issue, propose guided exploration-based deep Q-Network (GE-DQN) algorithm utilizing fixed advantage policy, which has faster learning speed compared classical Q-Network. Moreover, use shared decision module followed several network branches transform structure GE-DQN Branching Dueling (GE-BDQN) algorithm. Since branch GE-BDQN can decompose high-dimensional action, reduce approximate exponential growth number output neurons with increase sensors linear GE-DQN, ensuring applicability under large-scale From simulation results, it be found that proposed two achieve better advanced algorithms, up 36% performance gain.

Language: Английский

Resource Allocation and Offloading Strategy for UAV-Assisted LEO Satellite Edge Computing DOI Creative Commons
Hongxia Zhang,

Shiyu Xi,

Hongzhao Jiang

et al.

Drones, Journal Year: 2023, Volume and Issue: 7(6), P. 383 - 383

Published: June 7, 2023

In emergency situations, such as earthquakes, landslides and other natural disasters, the terrestrial communications infrastructure is severely disrupted unable to provide services IoT devices. However, tasks in scenarios often require high levels of computing power energy supply that cannot be processed quickly enough by devices locally computational offloading. addition, offloading server-equipped edge base stations may not always feasible due lack or distance. Since Low Orbit Satellites (LEO) have abundant resources, Unmanned Aerial Vehicles (UAVs) flexible deployment, LEO satellite servers via UAVs becomes straightforward, which provides ground-based Therefore, this paper investigates resource allocation a UAV-assisted multi-layer network, taking into account resources device task volumes. order minimise weighted sum consumption delay system, problem formulated constrained optimisation problem, then transformed Markov Decision Problem (MDP). We propose airspace integration network architecture, Deep Deterministic Policy Gradient Long short-term memory (DDPG-LSTM)-based algorithm solve problem. Simulation results demonstrate solution outperforms baseline approach our framework potential reliable communication situations.

Language: Английский

Citations

41
Aerial Edge Computing: A Survey DOI

Qinglou Zhang,

Ying Luo, Hong Jiang

et al.

IEEE Internet of Things Journal, Journal Year: 2023, Volume and Issue: 10(16), P. 14357 - 14374

Published: March 30, 2023

In the beyond 5G/6G era, aerial edge computing (AEC) is expected to be used as significant components in Internet of Things. AEC brings flexible deployment with Line-of-Sight communication links for task offloading and services. this article, we present a comprehensive survey technology. First, introduce three-layer architecture AEC, which includes satellite, unmanned aircraft vehicles, ground terminals. Second, illustrate challenges summarize recent studies terms performance metrics that include energy efficiency, latency, operation cost address AEC. Further, advanced technologies applied management, e.g., artificial intelligence (AI) distributed optimization. Finally, applications open issues are summarized classified study.

Language: Английский

Citations

24
DDL: Empowering Delivery Drones with Large-scale Urban Sensing Capability DOI
Xuecheng Chen, Haoyang Wang, Yuhan Cheng

et al.

IEEE Journal of Selected Topics in Signal Processing, Journal Year: 2024, Volume and Issue: 18(3), P. 502 - 515

Published: April 1, 2024

Language: Английский

Citations

10
A Survey on Applications of Unmanned Aerial Vehicles Using Machine Learning DOI Creative Commons
Karolayne Teixeira, Geovane Miguel, Hugerles S. Silva

et al.

IEEE Access, Journal Year: 2023, Volume and Issue: 11, P. 117582 - 117621

Published: Jan. 1, 2023

Unmanned Aerial Vehicles (UAVs) play an important role in many applications, including health, transport, telecommunications and safe rescue operations. Their adoption can improve the speed precision of applications when compared to traditional solutions based on handwork. The use UAVs brings scientific technological challenges. In this context, Machine Learning (ML) techniques provide several problems concerning civil military applications. An increasing number papers ML context have been published academic journals. work, we present a literature review UAVs, outlining most recurrent areas commonly used UAV results reveal that environment, communication security are among main research topics.

Language: Английский

Citations

18
Energy Consumption Modeling and Optimization of UAV-Assisted MEC Networks Using Deep Reinforcement Learning DOI
Ming Yan, Litong Zhang, Wei Jiang

et al.

IEEE Sensors Journal, Journal Year: 2024, Volume and Issue: 24(8), P. 13629 - 13639

Published: March 7, 2024

Unmanned aerial vehicle (UAV)-assisted multiaccess edge computing (MEC) technology has garnered significant attention and been successfully implemented in specific scenarios. The optimization of the network energy consumption relevant scenarios is essential for whole system performance due to constrained capacity UAVs. However, dynamic changes MEC resources make a challenge. In this article, multi-UAV-multiuser model established assess consumption, problem multi-UAV cooperation strategies formulated based on model. Then, multiagent deep deterministic policy gradient (MADDPG) algorithm reinforcement learning (DRL) employed resolve above problem. Each UAV equivalent single agent that cooperates with other agents train actors critic evaluation networks accomplish collaborative decision-making. addition, prioritized experience replay (PER) scheme used improve convergence training process. Simulation results show impact different by comparing algorithms. findings presented article serve as valuable reference future work optimization, specifically terms efficiency.

Language: Английский

Citations

8
ESUAV-NI: Endogenous Security Framework for UAV Perception System Based on Neural Immunity DOI
Heqing Li, Xinde Li, Zhentong Zhang

et al.

IEEE Transactions on Industrial Informatics, Journal Year: 2023, Volume and Issue: 20(1), P. 732 - 743

Published: April 28, 2023

Unmanned aerial vehicles (UAVs) represent an essential component of advanced intelligent equipment that can be used as perception system by installing various sensors such vision, hearing, touch, taste, and smell to achieve intelligently integrated environments. However, these with environmental information may threatened internal external attacks, causing a great challenge the security UAV. The original relied on expert knowledge base prevent but weaknesses lacking proactivity flexibility are gradually exposed. strong resistance survivability biological systems fill this capability gap provide new ideas for UAV system. Therefore, endogenous framework (ESUAV-NI) based neural immune is proposed in article. Through breeding artificial intelligence (AI) vaccines distributed hierarchical control, we protection Moreover, evaluated AI vaccine approach ESUAV-NI conducting extensive experiments threats imagery camouflage data, respectively. results show has superior performance

Language: Английский

Citations

14
Robust Trajectory and Offloading for Energy-Efficient UAV Edge Computing in Industrial Internet of Things DOI
Xiao Tang, Hongrui Zhang, Ruonan Zhang

et al.

IEEE Transactions on Industrial Informatics, Journal Year: 2023, Volume and Issue: 20(1), P. 38 - 49

Published: March 13, 2023

Efficient data processing and computation are essential for the Industrial Internet of Things (IIoT) to empower various applications, which can be significantly bottlenecked by limited energy capacity capability IIoT nodes. In this article, we employ an unmanned aerial vehicle (UAV) as edge server assist processing, while considering practical issue UAV jittering. Specifically, propose a joint design on trajectory offloading strategies minimize consumption due local computation, well transmission. We particularly address jittering that induces Gaussian-distributed uncertainties associated with flying waypoints, resulting in probabilistic-form speed constraints. exploit Bernstein-type inequality reformulate constraints deterministic forms decompose minimization solve separately within alternating optimization framework. The subproblems then tackled successive convex approximation technique. Simulation results show our proposal strictly guarantees robustness under effectively reduces compared baselines.

Language: Английский

Citations

13
Blockchain-enabled trust management for secure content caching in mobile edge computing using deep reinforcement learning DOI
Soumaya Bounaira, Ahmed Alioua, İsmahane Souici

et al.

Internet of Things, Journal Year: 2024, Volume and Issue: 25, P. 101081 - 101081

Published: Jan. 22, 2024

Language: Английский

Citations

5
Virtual Network Embedding for Task Offloading in IIoT: A DRL-Assisted Federated Learning Scheme DOI
Sheng Wu, Ning Chen, Guanghui Wen

et al.

IEEE Transactions on Industrial Informatics, Journal Year: 2024, Volume and Issue: 20(4), P. 6814 - 6824

Published: Jan. 24, 2024

The Industrial Internet of Things (IIoT) promotes the deep integration new-generation communication technologies and industrial ecology. However, popularity computing proliferation equipment scale make it a meaningful challenge to provide reasonable resource allocation for task offloading. Therefore, this article proposes novel two-stage coordinated, distributed, online multidomain virtual network embedding algorithm based on reinforcement learning (DRL)-assisted federated (FL) offloading in IIoT. We model IIoT as dynamic structure deploy local DRL servers each factory domain combined with distributed paradigm FL reduce fragmentation. Through global cooperation, environment is controlled fine macroscopic manner. In addition, mechanisms ensure privacy participant data. Finally, comprehensive evaluation demonstrates clear superiority proposed algorithm, which improves long-term revenue, utilization, success rate by average 17.66%, 5.97%, 4.52% compared baselines, respectively.

Language: Английский

Citations

5
UAV-Enabled Secure Communications via Collaborative Beamforming With Imperfect Eavesdropper Information DOI
Geng Sun, Xiaoya Zheng, Zemin Sun

et al.

IEEE Transactions on Mobile Computing, Journal Year: 2023, Volume and Issue: 23(4), P. 3291 - 3308

Published: May 5, 2023

Unmanned aerial vehicles (UAVs) are playing a pivotal role in wireless networks due to their high mobility and on-demand deployment advantages. However, the UAV-enabled communications susceptible be wiretapped by eavesdroppers strong line-of-sight (LoS) dominated air-ground channel. In this paper, we consider secure communication scenario, which group of UAVs form virtual antenna array (UVAA) transmit information towards remote base stations (BSs) via collaborative beamforming (CB), while multiple known unknown aiming wiretap information. Specifically, multi-objective optimization problem (SCMOP) is formulated achieve maximization worst-case secrecy rate, minimization maximum sidelobe level (SLL) as well flight energy consumption obtaining optimal locations excitation current weights concerning determining an receiver BS that can superior performance. To solve SCMOP demonstrated non-convex NP-hard, improved salp swarm algorithm (IMSSA) with several specific operating factors proposed. Simulations results demonstrate proposed IMSSA deal effectively outperforms other benchmark strategies. Moreover, multi-hop relay introduced verify reasonability UVAA system, two schemes necessity SCMOP. addition, performance system under certain unexpected circumstances estimated. Finally, experimental implementation conducted using Raspberry Pi practicality CB-based approach real-world scenarios.

Language: Английский

Citations

11