A sub-action aided deep reinforcement learning framework for latency-sensitive network slicing

Xiao, Da; Chen, Shuo; Ni, Wei; Zhang, Jie; Zhang, Andrew; Liu, Ren Ping

doi:10.1016/j.comnet.2022.109279

Cited by 6 publications

(9 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some papers [22,23,24,25] provide techno-economic analysis of slice allocation of MEC-enabled network, and find the trade-off between resource usage, service performances, cost and revenue. Other papers [26,19] also provide a techno-economic analysis of a similar problem but do not consider the MEC system.…”

Section: Related Workmentioning

confidence: 99%

“…Other papers [26,19] also provide a techno-economic analysis of a similar problem but do not consider the MEC system. Studies in [22,26,23] evaluate the slice admission, and focus on maximizing the provider's revenue by selecting which slices needs to be admitted with SLA constraints. Studying resource admission is a critical aspect of network slicing.…”

Section: Related Workmentioning

confidence: 99%

“…Generally, studies have been carried out to evaluate RAN-slicing [26,30,31], or a high-layer perspective of slice that can fall into RAN-transport or RAN-transportcore slicing [33,24,25,19,27,28,29]. Based on the…”

Section: Related Workmentioning

confidence: 99%

“…The common approach of RAN-slicing is to model a network slice as a fluid entity that needs to decide the number of PRBs and the user association [26,30,31]. The common approach of high-layer isolation is to model the network slice as an SFC composed of VNFs [33,24,25,19,27,28,29]. The slice as SFC allocation problem is then further divided into SFC resource partition (e.g., deciding how many resource should be allocated to each VNF request [24,25,28]) and VNFs embedding (e.g., deciding which VNF should be allocated to which part of the infrastructure [22,23,27,29]).…”

Section: Journal Pre-proofmentioning

confidence: 99%

“…The common approach of high-layer isolation is to model the network slice as an SFC composed of VNFs [33,24,25,19,27,28,29]. The slice as SFC allocation problem is then further divided into SFC resource partition (e.g., deciding how many resource should be allocated to each VNF request [24,25,28]) and VNFs embedding (e.g., deciding which VNF should be allocated to which part of the infrastructure [22,23,27,29]). However, most slice allocation and embedding problems consider a physical infrastructure, which a single InP offers.…”

Section: Journal Pre-proofmentioning

confidence: 99%

See 4 more Smart Citations

Resource Allocation for Cost Minimization of a Slice Broker in a 5g-Mec Scenario

Sarah

Nencioni

2022

SSRN Journal

View full text Add to dashboard Cite

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Journal Pre-proofmentioning

confidence: 99%

Section: Journal Pre-proofmentioning

confidence: 99%

See 3 more Smart Citations

Resource Allocation for Cost Minimization of a Slice Broker in a 5g-Mec Scenario

Sarah

Nencioni

2022

SSRN Journal

View full text Add to dashboard Cite

Dynamic and efficient resource allocation for 5G end‐to‐end network slicing: A multi‐agent deep reinforcement learning approach

Asim Ejaz,

Wu,

Iqbal

2024

Int J Communication

View full text Add to dashboard Cite

SummaryThe rapid evolution of user equipment (UE) and 5G networks drives significant transformations, bringing technology closer to end‐users. Managing resources in densely crowded areas such as airports, train stations, and bus terminals poses challenges due to diverse user demands. Integrating mobile edge computing (MEC) and network function virtualization (NFV) becomes vital when the service provider's (SP) primary goal is maximizing profitability while maintaining service level agreement (SLA). Considering these challenges, our study addresses an online resource allocation problem in an MEC network where computing resources are limited, and the SP aims to boost profit by securely admitting more UE requests at each time slot. Each UE request arrival rate is unknown, and the requirement is specific resources with minimum cost and delay. The optimization problem objective is achieved by allocating resources to requests at the MEC network in appropriate cloudlets, utilizing abandoned instances, reutilizing idle and soft slice instances to shorten delay and reduce costs, and immediately scaling inappropriate instances, thus minimizing the instantiation of new instances. This paper proposes a deep reinforcement learning (DRL) method for request prediction and resource allocation to mitigate unnecessary resource waste. Simulation results demonstrate that the proposed approach effectively accepts network slice requests to maximize profit by leveraging resource availability, reutilizing instantiated resources, and upholding goodwill and SLA. Through extensive simulations, we show that our proposed DRL‐based approach outperforms other state‐of‐the‐art techniques, namely, MaxSR, DQN, and DDPG, by 76%, 33%, and 23%, respectively.

show abstract

Markov Decision Process and Artificial Neural Network for Resource Capacity Planning in 5G Network Slicing

Ghebrial,

Leonteva,

Kochetkova

et al. 2024

Communications in Computer and Information Science

View full text Add to dashboard Cite

A sub-action aided deep reinforcement learning framework for latency-sensitive network slicing

Cited by 6 publications

References 30 publications

Resource Allocation for Cost Minimization of a Slice Broker in a 5g-Mec Scenario

Resource Allocation for Cost Minimization of a Slice Broker in a 5g-Mec Scenario

Dynamic and efficient resource allocation for 5G end‐to‐end network slicing: A multi‐agent deep reinforcement learning approach

Markov Decision Process and Artificial Neural Network for Resource Capacity Planning in 5G Network Slicing

Contact Info

Product

Resources

About