Joint Optimization of Caching and Computation in Multi-Server NOMA-MEC System via Reinforcement Learning

Li, Shilu; Li, Baogang; Zhao, Wei

doi:10.1109/access.2020.3002895

Cited by 40 publications

(25 citation statements)

References 34 publications

(40 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…1 Edge computing scenarios in 5G Industrial Internet of Things environment allocation was proposed in [20]. A deep Q-network for multi-agent settings (MADQN) based on the predicted popularity was introduced to solve the caching and offloading problems [21].…”

Section: Computation Offloadingmentioning

confidence: 99%

Efficient task offloading using particle swarm optimization algorithm in edge computing for industrial internet of things

You

Tang

2021

J Cloud Comp

View full text Add to dashboard Cite

As a new form of computing based on the core technology of cloud computing and built on edge infrastructure, edge computing can handle computing-intensive and delay-sensitive tasks. In mobile edge computing (MEC) assisted by 5G technology, offloading computing tasks of edge devices to the edge servers in edge network can effectively reduce delay. Designing a reasonable task offloading strategy in a resource-constrained multi-user and multi-MEC system to meet users’ needs is a challenge issue. In industrial internet of things (IIoT) environment, considering the rapid increase of industrial edge devices and the heterogenous edge servers, a particle swarm optimization (PSO)-based task offloading strategy is proposed to offload tasks from resource-constrained edge devices to edge servers with energy efficiency and low delay style. A multi-objective optimization problem that considers time delay, energy consumption and task execution cost is proposed. The fitness function of the particle represents the total cost of offloading all tasks to different MEC servers. The offloading strategy based on PSO is compared with the genetic algorithm (GA) and the simulated annealing algorithm (SA) through simulation experiments. The experimental results show that the task offloading strategy based on PSO can reduce the delay of the MEC server, balance the energy consumption of the MEC server, and effectively realize the reasonable resource allocation.

show abstract

Section: Computation Offloadingmentioning

confidence: 99%

Efficient task offloading using particle swarm optimization algorithm in edge computing for industrial internet of things

You

Tang

2021

J Cloud Comp

View full text Add to dashboard Cite

show abstract

“…6 Games enable agents to achieve the goals of games, such as winning the games or increasing human players' engagement. In [63], two players select their movements (i.e., up or down) in a pong game. MADRL addresses the challenge of high dynamicity using a common reward received after winning a game to encourage cooperative behavior among the players.…”

Section: Applicationsmentioning

confidence: 99%

“…In [63], Shilu et al embedded SADRL in mobile nodes to offload compute-intensive tasks (A.5) to a centralized entity (i.e., the base station) (X.3.1) based on their computational resources and the delay incurred in traffic offload. The mobile nodes maximize the global reward based on the shared information from base stations in a collaborative (X.2.2) manner.…”

Section: Shilu's Sadrl Approach In a Multi-agent Environmentmentioning

confidence: 99%

Applications of Multi-Agent Deep Reinforcement Learning: Models and Algorithms

et al. 2021

View full text Add to dashboard Cite

Recent advancements in deep reinforcement learning (DRL) have led to its application in multi-agent scenarios to solve complex real-world problems, such as network resource allocation and sharing, network routing, and traffic signal controls. Multi-agent DRL (MADRL) enables multiple agents to interact with each other and with their operating environment, and learn without the need for external critics (or teachers), thereby solving complex problems. Significant performance enhancements brought about by the use of MADRL have been reported in multi-agent domains; for instance, it has been shown to provide higher quality of service (QoS) in network resource allocation and sharing. This paper presents a survey of MADRL models that have been proposed for various kinds of multi-agent domains, in a taxonomic approach that highlights various aspects of MADRL models and applications, including objectives, characteristics, challenges, applications, and performance measures. Furthermore, we present open issues and future directions of MADRL.

show abstract

“…Further, a weighted double DQN scheme is utilized for avoiding overestimation of Q value. Reference [96] applies an RNN to predict content popularity by collecting historical requests and the output represents the popularity in the near future. Then, the prediction is employed for cooperative caching and computation offloading among MEC servers, which is modeled as an ILP problem.…”

Section: Drlmentioning

confidence: 99%

A Survey of Deep Learning for Data Caching in Edge Network

Wang

Friderikos

2020

Informatics

View full text Add to dashboard Cite

The concept of edge caching provision in emerging 5G and beyond mobile networks is a promising method to deal both with the traffic congestion problem in the core network, as well as reducing latency to access popular content. In that respect, end user demand for popular content can be satisfied by proactively caching it at the network edge, i.e., at close proximity to the users. In addition to model-based caching schemes, learning-based edge caching optimizations have recently attracted significant attention, and the aim hereafter is to capture these recent advances for both model-based and data-driven techniques in the area of proactive caching. This paper summarizes the utilization of deep learning for data caching in edge network. We first outline the typical research topics in content caching and formulate a taxonomy based on network hierarchical structure. Then, many key types of deep learning algorithms are presented, ranging from supervised learning to unsupervised learning, as well as reinforcement learning. Furthermore, a comparison of state-of-the-art literature is provided from the aspects of caching topics and deep learning methods. Finally, we discuss research challenges and future directions of applying deep learning for caching.

show abstract

Joint Optimization of Caching and Computation in Multi-Server NOMA-MEC System via Reinforcement Learning

Cited by 40 publications

References 34 publications

Efficient task offloading using particle swarm optimization algorithm in edge computing for industrial internet of things

Efficient task offloading using particle swarm optimization algorithm in edge computing for industrial internet of things

Applications of Multi-Agent Deep Reinforcement Learning: Models and Algorithms

A Survey of Deep Learning for Data Caching in Edge Network

Contact Info

Product

Resources

About