Joint optimization of energy consumption and time delay in IoT-fog-cloud computing environments using NSGA-II metaheuristic algorithm

Journal of Ambient Intelligence and Humanized Computing, Journal Year: 2021, Volume and Issue: 14(3), P. 1675 - 1698

Published: July 23, 2021

Language: Английский

---

Integration of D2D, Network Slicing, and MEC in 5G Cellular Networks: Survey and Challenges

IEEE Access, Journal Year: 2021, Volume and Issue: 9, P. 37590 - 37612

Published: Jan. 1, 2021

With the tremendous demand for connectivity anywhere and anytime, existing network architectures should be modified. To cope with challenges that arise due to increasing flood of devices/users a diverse range application requirements, new technologies concepts must integrated enable their benefits. Service providers business companies are looking areas research enhance overall system performance. This article gives detailed survey about recent 5G technologies, solutions they provide, effect caused by addition current cellular networks. It is based on three most important concepts: Device (D2D), Network Slicing (NS), Mobile Edge Computing (MEC). study proposes design future networks integration all technologies. believed spectrum efficiency, energy throughput will greatly improved using D2D. The delay computational load reduced as tasks handled edge routers located at base stations. Thus offloading core capital expenses operational significantly slicing network.

Language: Английский

---

Towards effective offloading mechanisms in fog computing

Multimedia Tools and Applications, Journal Year: 2021, Volume and Issue: 81(2), P. 1997 - 2042

Published: Oct. 19, 2021

Language: Английский

---

Task offloading in fog computing: A survey of algorithms and optimization techniques

Computer Networks, Journal Year: 2022, Volume and Issue: 214, P. 109137 - 109137

Published: July 3, 2022

Language: Английский

---

Blockchain Systems, Technologies, and Applications: A Methodology Perspective

IEEE Communications Surveys & Tutorials, Journal Year: 2022, Volume and Issue: 25(1), P. 353 - 385

Published: Sept. 6, 2022

In the past decade, blockchain has shown a promising vision to build trust without any powerful third party in secure, decentralized and scalable manner. However, due wide application future development from cryptocurrency Internet of Things, is an extremely complex system enabling integration with mathematics, computer science, communication network engineering, etc. By revealing intrinsic relationship between communication, networking computing methodological perspective, it provided view challenge that engineers, experts researchers hardly fully understand process systematic top bottom. this article we first introduce how works, research activities challenges, illustrate roadmap involving classic methodologies typical use cases topics. Second, systems, adopt stochastic process, game theory, optimization machine learning study running processes design protocols/algorithms are discussed details. Moreover, advantages limitations using these methods also summarized as guide work be further considered. Finally, some remaining problems technical, commercial political views open issues. The main findings will provide survey perspective theoretical model for fundamentals understanding, service blockchain-based mechanisms algorithms, well apply

Language: Английский

---

Machine and Deep Learning for Resource Allocation in Multi-Access Edge Computing: A Survey

IEEE Communications Surveys & Tutorials, Journal Year: 2022, Volume and Issue: 24(4), P. 2449 - 2494

Published: Jan. 1, 2022

With the rapid development of Internet-of-Things (IoT) devices and mobile communication technologies, Multi-access Edge Computing (MEC) has emerged as a promising paradigm to extend cloud computing storage capabilities edge cellular networks, near IoT devices. MEC enables with limited battery capacity computation/storage execute their computation-intensive latency-sensitive applications at networks. However, efficiently these in systems, each task must be properly offloaded scheduled onto servers. Additionally, servers may intelligently balance share resources satisfy application QoS QoE. Therefore, effective resource allocation (RA) mechanisms are vital for ensuring its foreseen advantages. Recently, Machine Learning (ML) Deep (DL) have key methods many challenging aspects MEC. Particularly, ML DL play crucial role addressing challenges RA This paper presents comprehensive survey ML/DL-based We first present tutorials that demonstrate advantages applying Then, we enabling technologies quickly running ML/DL training inference Afterward, provide an in-depth recent works used from three aspects: (1) offloading; (2) scheduling; (3) joint allocation. Finally, discuss future research directions

Language: Английский

---

Mobile edge computing for V2X architectures and applications: A survey

Computer Networks, Journal Year: 2022, Volume and Issue: 206, P. 108797 - 108797

Published: Feb. 2, 2022

Language: Английский

---

A Survey on Mobile Edge Computing for Video Streaming: Opportunities and Challenges

IEEE Access, Journal Year: 2022, Volume and Issue: 10, P. 120514 - 120550

Published: Jan. 1, 2022

5G communication brings substantial improvements in the quality of service provided to various applications by achieving higher throughput and lower latency. However, interactive multimedia (e.g., ultra high definition video conferencing, 3D multiview streaming, crowd-sourced cloud gaming, virtual augmented reality) are becoming more ambitious with volume low latency streams putting strict demands on already congested networks. Mobile Edge Computing (MEC) is an emerging paradigm that extends computing capabilities edge network i.e., at base station level. To meet requirements avoid end-to-end remote data centers, MEC allows store process content caching, transcoding, pre-processing) stations. Both demand live streaming can utilize improve existing services develop novel use cases, such as analytics, targeted advertisements. expected reshape future providing ultra-reliable reality, autonomous vehicles), pervasive real-time analytics), blockchain-enabled architecture for secure streaming. This paper presents a comprehensive survey recent developments MEC-enabled bringing unprecedented improvement enable cases. A detailed review state-of-the-art presented covering caching schemes, optimal computation offloading, cooperative offloading artificial intelligence (i.e., machine learning, deep reinforcement learning) MEC-assisted services.

Language: Английский

---

Delay-Optimal Task Offloading for UAV-Enabled Edge-Cloud Computing Systems

IEEE Access, Journal Year: 2022, Volume and Issue: 10, P. 51575 - 51586

Published: Jan. 1, 2022

The emergence of delay-sensitive and computationally-intensive mobile applications services pose a significant challenge for Unmanned Aerial Vehicles (UAVs) devices due to the scarcity in their resources such as computational power battery lifetime. Mobile cloud computing has been introduced promising solution overcome these limitations through task offloading. However, high-latency security issues are considered main challenges this paradigm. Subsequently, edge-cloud paradigm widely used help mitigate issues. Nevertheless, current offloading models permit UAVs execute intensive tasks at connected edge server, which leads excessive loads large number thereby increases delay. Therefore, paper, we propose delay-optimal approach multi-tier multi-user environment. problem is formulated an optimization model using Integer Linear Programming (ILP) techniques minimize total service time UAVs. Simulation results demonstrate that proposed not only saves by 33.5% 55% execution policies respectively, but also scales well

Language: Английский

---

A comprehensive survey on reinforcement-learning-based computation offloading techniques in Edge Computing Systems

Journal of Network and Computer Applications, Journal Year: 2023, Volume and Issue: 216, P. 103669 - 103669

Published: May 9, 2023

In recent years, the number of embedded computing devices connected to Internet has exponentially increased. At same time, new applications are becoming more complex and computationally demanding, which can be a problem for devices, especially when they battery powered. this context, concepts computation offloading edge computing, allow fully or partially offloaded executed on servers close in network, have arisen received increasing attention. Then, design algorithms make decision tasks should offloaded, where execute them, is crucial. One options that been gaining momentum lately use Reinforcement Learning (RL) and, particular, Deep (DRL), enables learning optimal near-optimal policies adapted each particular scenario. Although RL techniques solve systems covered by some surveys, it done limited way. For example, surveys analysed various networking problems, with being one but not primary focus. Other other hand, reviewed problem, just approaches considered. To best our knowledge, first survey specifically focuses DRL system. We present comprehensive detailed survey, we analyse classify research papers terms cases, network architectures, objectives, algorithms, decision-making approaches, time-varying characteristics considered scenarios. include series tables help researchers identify relevant based specific features, scenarios most frequently literature. Finally, identifies challenges, future directions areas further study.

Language: Английский

---