Next-generation predictive maintenance: leveraging blockchain and dynamic deep learning in a domain-independent system DOI Creative Commons
Montdher Alabadi, Adib Habbal

PeerJ Computer Science, Journal Year: 2023, Volume and Issue: 9, P. e1712 - e1712

Published: Dec. 6, 2023

The fourth industrial revolution, often referred to as Industry 4.0, has revolutionized the manufacturing sector by integrating emerging technologies such artificial intelligence (AI), machine and deep learning, Industrial Internet of Things (IIoT), cloud computing, cyber physical systems (CPSs) cognitive throughout production life cycle. Predictive maintenance (PdM) emerges a critical component, utilizing data analytic track health proactively detect machinery failures. Deep learning (DL), is pivotal in this context, offering superior accuracy prediction through neural networks’ processing capabilities. However, DL adoption PdM faces challenges, including continuous model updates domain dependence. Meanwhile, centralized models, prevalent PdM, pose security risks central points failure unauthorized access. To address these issues, study presents an innovative decentralized system DL, blockchain, storage based on InterPlanetary File System (IPFS) for accurately predicting Remaining Useful Lifetime (RUL). handles predictive tasks, while blockchain secures orchestration. Decentralized safeguards metadata training dynamic models. features synchronized two pipelines time series data, encompassing mechanisms. detailed material methods research shed light system’s development validation processes. Rigorous confirms accuracy, performance, experimental testbed. results demonstrate updating independence. Prediction surpass state-of-the-art models terms root mean squared error (RMSE) score. Blockchain-based scalability performance was tested smart contract gas usage, analysis shows efficient across varying input output scales. A comprehensive CIA highlights robust features, addressing confidentiality, integrity, availability aspects. proposed system, which incorporates technology, storage, potential improve overcome significant obstacles. Consequently, holds promising implications advancement context 4.0.

Language: Английский

A review of IoT applications in healthcare DOI Creative Commons
Chunyan Li, Jiaji Wang, Shuihua Wang‎

et al.

Neurocomputing, Journal Year: 2023, Volume and Issue: 565, P. 127017 - 127017

Published: Nov. 9, 2023

Integrating Internet of Things (IoT) technologies in the healthcare industry represents a transformative shift with tangible benefits. This paper provides detailed examination IoT adoption healthcare, focusing on specific sensor types and communication methods. It underscores successful real-world applications, including remote patient monitoring, individualized treatment strategies, streamlined delivery. Furthermore, it delves into intricate challenges to realizing full potential healthcare. includes addressing data security concerns, ensuring seamless interoperability, optimizing use IoT-generated data. The seeks inspire practitioners researchers by highlighting practical implications emphasizing ways can enhance care, resource allocation, overall efficiency.

Language: Английский

Citations

94
Smart performance optimization of energy‐aware scheduling model for resource sharing in 5G green communication systems DOI Creative Commons

Sivakumar Sangeetha,

J. Logeshwaran, Muhammad Faheem

et al.

The Journal of Engineering, Journal Year: 2024, Volume and Issue: 2024(2)

Published: Feb. 1, 2024

Abstract This paper presents an analysis of the performance Energy Aware Scheduling Algorithm (EASA) in a 5G green communication system. systems rely on EASA to manage resource sharing. The aim proposed model is improve efficiency and energy consumption sharing systems. main objective address challenges achieving optimal utilization minimizing these To achieve this goal, study proposes novel energy‐aware scheduling that takes into consideration specific characteristics incorporates intelligent techniques for optimizing allocation decisions, while also considering constraints. methodology used involves combination mathematical simulation studies. formulate optimization problem design model, simulations are evaluate its various scenarios. EASM reached 91.58% false discovery rate, 64.33% omission 90.62% prevalence threshold, 91.23% critical success index. results demonstrate effectiveness terms reducing maintaining high level utilization.

Language: Английский

Citations

19
A Review of Urban Digital Twins Integration, Challenges, and Future Directions in Smart City Development DOI Open Access
Silvia Mazzetto

Sustainability, Journal Year: 2024, Volume and Issue: 16(19), P. 8337 - 8337

Published: Sept. 25, 2024

This review paper explores Urban Digital Twins (UDTs) and their crucial role in developing smarter cities, focusing on making urban areas more sustainable well-planned. The methodology adopted an extensive literature across multiple academic databases related to UDTs smart sustainability, environments, conducted by a bibliometric analysis using VOSviewer identify key research trends qualitative through thematic categorization. shows how can significantly change cities are managed planned examining examples from like Singapore Dubai. study points out the main hurdles gathering data, connecting systems, handling vast amounts of information, different technologies work together. It also sheds light what is missing current research, such as need for solid rules effectively, better cooperation between various city deeper look into affect society. To address gaps, this highlights necessity interdisciplinary collaboration. calls establishing comprehensive models, universal standards, comparative studies among traditional UDT methods. Finally, it encourages industry, policymakers, academics join forces realizing sustainable, cities.

Language: Английский

Citations

18
Task Allocation Methods and Optimization Techniques in Edge Computing: A Systematic Review of the Literature DOI Creative Commons
Vasilios Patsias, Petros Amanatidis, Dimitris Karampatzakis

et al.

Future Internet, Journal Year: 2023, Volume and Issue: 15(8), P. 254 - 254

Published: July 28, 2023

Task allocation in edge computing refers to the process of distributing tasks among various nodes an network. The main challenges task include determining optimal location for each based on requirements such as processing power, storage, and network bandwidth, adapting dynamic nature Different approaches centralized, decentralized, hybrid, machine learning algorithms. Each approach has its strengths weaknesses choice will depend specific application. In more detail, selection most methods depends architecture configuration type, like mobile (MEC), cloud-edge, fog computing, peer-to-peer etc. Thus, is a complex, diverse, challenging problem that requires balance trade-offs between multiple conflicting objectives energy efficiency, data privacy, security, latency, quality service (QoS). Recently, increased number research studies have emerged regarding performance evaluation optimization devices. While several survey articles described current state-of-the-art methods, this work focuses comparing contrasting different algorithms, well types are frequently used systems.

Language: Английский

Citations

25
Dynamic Priority-Based Task Scheduling and Adaptive Resource Allocation Algorithms for Efficient Edge Computing in Healthcare Systems DOI Creative Commons

J. Anand,

B. Karthikeyan

Results in Engineering, Journal Year: 2025, Volume and Issue: unknown, P. 104342 - 104342

Published: Feb. 1, 2025

Language: Английский

Citations

1
A Survey on IoT-Edge-Cloud Continuum Systems: Status, Challenges, Use Cases, and Open Issues DOI Creative Commons
Panagiotis K. Gkonis, Anastasios Giannopoulos, Panagiotis Trakadas

et al.

Future Internet, Journal Year: 2023, Volume and Issue: 15(12), P. 383 - 383

Published: Nov. 28, 2023

The rapid growth in the number of interconnected devices on Internet (referred to as Things—IoT), along with huge volume data that are exchanged and processed, has created a new landscape network design operation. Due limited battery size computational capabilities IoT nodes, processing usually takes place external devices. Since latency minimization is key concept modern-era networks, edge servers close proximity nodes gather process related data, while some cases offloading cloud might have take place. interconnection vast heterogeneous cloud, where IoT, edge, converge form computing continuum, also known IoT-edge-cloud (IEC) continuum. Several challenges associated this systems’ architectural approach, including (i) connection programming protocols aimed at properly manipulating over diverse infrastructures; (ii) efficient task algorithms optimizing services execution; (iii) support for security privacy enhancements during transfer deal existent even unforeseen attacks threats landscape; (iv) scalability, flexibility, reliability guarantees face expected mobility systems; (v) optimal resource allocation mechanisms make most out available resources. These will become more significant towards era sixth-generation (6G) which be based integration various cutting-edge technologies. Therefore, goal survey paper present all recent developments field IEC continuum systems, respect aforementioned deployment challenges. In same context, potential limitations future highlighted well. Finally, indicative use presented from an perspective.

Language: Английский

Citations

17
Decentralized Replica Management in Latency-Bound Edge Environments for Resource Usage Minimization DOI Creative Commons
Luca Ferrucci, Matteo Mordacchini, Patrizio Dazzi

et al.

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 19229 - 19249

Published: Jan. 1, 2024

The Internet is experiencing a fast expansion at its edges. wide availability of heterogeneous resources the Edge pivotal in definition and extension traditional Cloud solutions toward supporting development new applications. However, dynamic distributed nature these poses challenges for optimization behavior system. New decentralized self-organizing methods are needed to face Cloud-Edge scenario's needs optimize exploitation resources. In this paper we propose adaptive solution that reduces number replicas application services executed throughout system, all while ensuring latency constraints applications met, thus allowing also meet end users' QoS requirements. Experimental evaluations through simulation show effectiveness proposed approach.

Language: Английский

Citations

6
Resource optimization in edge and SDN-based edge computing: a comprehensive study DOI

Ajay Nain,

Sophiya Sheikh, Mohammad Shahid

et al.

Cluster Computing, Journal Year: 2024, Volume and Issue: 27(5), P. 5517 - 5545

Published: Feb. 8, 2024

Language: Английский

Citations

6
Forwarding Strategies for Named Data Networking Based IoT: Requirements, Taxonomy, and Open Research Challenges DOI Creative Commons
Naeem Ali Askar, Adib Habbal, Feras Zen Alden

et al.

IEEE Access, Journal Year: 2023, Volume and Issue: 11, P. 78363 - 78383

Published: Jan. 1, 2023

The Internet of Things (IoT) aims to efficiently connect various entities, including humans, machines, smart devices, physical environments, and others, so they can communicate exchange data in real time. However, due the massive amount transferred, presence devices with limited resources, heterogeneity, mobility support would make it difficult create a robust network respect performance an IoT context. In order disseminate enormous volume automated data, Named Data Networking (NDN), viable networking design for future Internet, has been proposed. NDN shown great potential because built-in naming, caching, mobility, security. Forwarding strategies play important role successful deployment NDN-based IoT. this article, we introduce forwarding emphasizing on characteristics requirements. We classify then discuss detail certain exemplary schemes. Additionally, compare several aspects current methods that are now use, types strategy, particular issues, type solution, assessment metrics, simulation platform. wrap up our contribution by outlining major open research issues guide investigations area. anticipate survey will help community researchers' understanding environments.

Language: Английский

Citations

13
Optimized CNN Architectures Benchmarking in Hardware-Constrained Edge Devices in IoT Environments DOI Creative Commons
Paúl D. Rosero-Montalvo, Pınar Tözün, Wilmar Hernández

et al.

IEEE Internet of Things Journal, Journal Year: 2024, Volume and Issue: 11(11), P. 20357 - 20366

Published: Feb. 23, 2024

Internet of Things (IoT) and Edge devices have grown in their application fields due to Machine learning (ML) models capacity classify images into previously known labels, working close the end-user. However, model might be trained with several convolutional neural network (CNN) architectures that can affect its performance when developed hardware-constrained environments, such as: devices. In addition, new training trends suggest using transfer techniques get an excellent feature extractor obtained from one domain use it a domain, which has not enough train whole model. light these trends, this work benchmarks most representative CNN on emerging devices, some hardware accelerators. The ML were optimized small set IoT environments learning. Our results show unfreezing until last 20 layers model's architecture fine-tuned correctly depending architecture. Additionally, quantization is suitable optimization technique shrink 2x or 3x times leading lighter memory footprint, lower execution time, battery consumption. Finally, Coral Dev Board boost 100x inference process, EfficientNet keeps same classification accuracy even adopted environment.

Language: Английский

Citations

5