Edge Computing Resource Allocation Algorithm for NB-IoT Based on Deep Reinforcement Learning

CHU, Jiawen; PAN, Chunyun; WANG, Yafei; YUN, Xiang; Li, Xuehua

doi:10.1587/transcom.2022ebp3076

Cited by 2 publications

(2 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Zheng et al [30] investigated the computation offloading problem in MEC systems under user mobility and task randomness environments, aiming to minimize energy consumption under delay and finite computing capacity constraints. Tokuda et al [31] proposed a MEC server resource allocation model to handle changes in user population, considering the fluctuation in the number of users caused by user mobility in MEC systems. He et al [32] proposed a novel ultra-dense network architecture based on MEC, where each MEC server is associated with a cluster of user-centric access points, acting as mobile agents.…”

Section: User Mobilitymentioning

confidence: 99%

Particle Swarm Optimization with Latin Hypercube Sampling for Mobility-Aware Task Offloading in D2D-Assisted MEC Networks

Song,

Tao,

Gao

et al. 2024

Preprint

View full text Add to dashboard Cite

In this paper, we consider a Device-to-Device (D2D)-assisted Mobile Edge Computing (MEC) system in which the edge server collaboratively performs tasks with users who possess mobile characteristics. The computational resources of the edge server and the users are limited. Our objective is to minimize latency and energy consumption while maximizing system stability by determining the offloading decisions for all tasks. However, solving this problem is particularly challenging due to the coupling between offloading decisions and user mobility characteristics. To address this challenge, a Particle Swarm Optimization (PSO) algorithm based on Latin Hypercube Sampling (LHS) is proposed. PSO can be employed for iterative computation to obtain the optimal solution by treating the offloading decisions for all tasks as particles. Furthermore, employing LHS for the initialization and updating of the particle swarm can significantly enhance the performance and convergence of the PSO. Simulation results indicate that our proposed algorithm outperforms other algorithms significantly in terms of latency, energy consumption, and system stability while also exhibiting excellent robustness.

show abstract

Section: User Mobilitymentioning

confidence: 99%

Particle Swarm Optimization with Latin Hypercube Sampling for Mobility-Aware Task Offloading in D2D-Assisted MEC Networks

Song,

Tao,

Gao

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…Moreover, in the field of Industrial Internet, terminal and edge devices have limited computing power. When dealing with complex nonconvex problems, traditional machine learning methods often encounter issues such as nonconvergence or local convergence [13], making the system optimization challenges even more complex. Additionally, factories place high importance on data privacy [14].…”

Section: Introductionmentioning

confidence: 99%

Federated Deep Reinforcement Learning for Energy-Efficient Edge Computing Offloading and Resource Allocation in Industrial Internet

Zhang

Pan

2023

Applied Sciences

View full text Add to dashboard Cite

Industrial Internet mobile edge computing (MEC) deploys edge servers near base stations to bring computing resources to the edge of industrial networks to meet the energy-saving requirements of Industrial Internet terminal devices. This paper considers a wireless MEC system in an intelligent factory that has multiple edge servers and mobile smart industrial terminal devices. In this paper, the terminal device has the choice of either offloading the task in whole or in part to the edge server, or performing it locally. Through combined optimization of the task offload ratio, number of subcarriers, transmission power, and computing frequency, the system can achieve minimum total energy consumption. A computing offloading and resource allocation approach that combines federated learning (FL) and deep reinforcement learning (DRL) is suggested to address the optimization problem. According to the simulation results, the proposed algorithm displays fast convergence. Compared with baseline algorithms, this algorithm has significant advantages in optimizing the performance of energy consumption.

show abstract

Edge Computing Resource Allocation Algorithm for NB-IoT Based on Deep Reinforcement Learning

Cited by 2 publications

References 22 publications

Particle Swarm Optimization with Latin Hypercube Sampling for Mobility-Aware Task Offloading in D2D-Assisted MEC Networks

Particle Swarm Optimization with Latin Hypercube Sampling for Mobility-Aware Task Offloading in D2D-Assisted MEC Networks

Federated Deep Reinforcement Learning for Energy-Efficient Edge Computing Offloading and Resource Allocation in Industrial Internet

Contact Info

Product

Resources

About