Deep Q-learning for 5G network slicing with diverse resource stipulations and dynamic data traffic

Network Slicing is a promising technology for providing customized logical and virtualized networks for the industry’s vertical segments.This paper proposes SARA and DSARA for the performance of admission control and resource allocation for network slice requests of eMBB, URLLC, and MIoT type in the 5G core network. SARA introduced a Q-learning based algorithm and DSARA a DQN-based algorithm to select the most profitable requests from a set that arrived in given time windows. These algorithms are model-free, meaning they do not make assumptions about the substrate network as do optimization based approaches.

Download Full-text

Admission Control for 5G Network Slicing based on (Deep) Reinforcement Learning

10.36227/techrxiv.14498190 ◽

2021 ◽

Author(s):

William Fernando Villota Jácome ◽

Oscar Mauricio Caicedo Rendon ◽

Nelson Luis Saldanha da Fonseca

Keyword(s):

Resource Allocation ◽

Reinforcement Learning ◽

Admission Control ◽

Time Windows ◽

Core Network ◽

Network Slicing ◽

Q Learning ◽

Model Free ◽

5G Network ◽

Substrate Network

Network Slicing is a promising technology for providing customized logical and virtualized networks for the industry’s vertical segments.This paper proposes SARA and DSARA for the performance of admission control and resource allocation for network slice requests of eMBB, URLLC, and MIoT type in the 5G core network. SARA introduced a Q-learning based algorithm and DSARA a DQN-based algorithm to select the most profitable requests from a set that arrived in given time windows. These algorithms are model-free, meaning they do not make assumptions about the substrate network as do optimization based approaches.

Download Full-text

Data Traffic Model in Machine to Machine Communications over 5G Network Slicing

2016 9th International Conference on Developments in eSystems Engineering (DeSE) ◽

10.1109/dese.2016.54 ◽

2016 ◽

Cited By ~ 4

Author(s):

Mohammed Dighriri ◽

Ali Saeed Dayem Alfoudi ◽

Gyu Myoung Lee ◽

Thar Baker

Keyword(s):

Traffic Model ◽

Data Traffic ◽

Network Slicing ◽

Machine To Machine ◽

5G Network

Download Full-text

5G Mobile Services and Scenarios: Challenges and Solutions

Sustainability ◽

10.3390/su10103626 ◽

2018 ◽

Vol 10 (10) ◽

pp. 3626 ◽

Cited By ~ 18

Author(s):

Yousaf Zikria ◽

Sung Kim ◽

Muhammad Afzal ◽

Haoxiang Wang ◽

Mubashir Rehmani

Keyword(s):

Mobile Network ◽

Future Internet ◽

Autonomous Driving ◽

Mobile Services ◽

Data Traffic ◽

Fifth Generation ◽

Wide Range ◽

Industrial Iot ◽

5G Network ◽

Massive Number

The Fifth generation (5G) network is projected to support large amount of data traffic and massive number of wireless connections. Different data traffic has different Quality of Service (QoS) requirements. 5G mobile network aims to address the limitations of previous cellular standards (i.e., 2G/3G/4G) and be a prospective key enabler for future Internet of Things (IoT). 5G networks support a wide range of applications such as smart home, autonomous driving, drone operations, health and mission critical applications, Industrial IoT (IIoT), and entertainment and multimedia. Based on end users’ experience, several 5G services are categorized into immersive 5G services, intelligent 5G services, omnipresent 5G services, autonomous 5G services, and public 5G services. In this paper, we present a brief overview of 5G technical scenarios. We then provide a brief overview of accepted papers in our Special Issue on 5G mobile services and scenarios. Finally, we conclude this paper.

Download Full-text