TY - JOUR
T1 - Intelligent Network Slicing for B5G and 6G
T2 - Resource Allocation, Service Provisioning, and Security
AU - Wang, Jiadai
AU - Li, Yuanhao
AU - Liu, Jiajia
AU - Kato, Nei
N1 - Publisher Copyright:
© 2002-2012 IEEE.
PY - 2024/6/1
Y1 - 2024/6/1
N2 - B5G and 6G are designed to achieve break-throughs in mobile networks, and are expected to have features such as seamless global coverage, full virtualization, and ubiquitous intelligence. These promising features will spawn lots of new verticals and applications, creating an urgent need for on-demand networks. At this point, network slicing has been recognized as an excellent solution to this challenge due to its advantages of flexible customization and logical isolation. Driven by massive data, intelligent network slicing is further endowed with capabilities of real-time perception, accurate prediction, and adaptive decision making, which brings unlimited potential for the development of B5G/6G. To this end, we investigate the application and prospect of intelligent slicing for B5G/6G in three areas, that is, resource allocation, service provisioning, and security. We also list the specific problems in these areas according to each slice lifecycle phase. In addition, a case study of using reinforcement learning for slice admission control is provided to improve the slice admission rate under limited resources and illustrate the advantages of intelligent network slicing.
AB - B5G and 6G are designed to achieve break-throughs in mobile networks, and are expected to have features such as seamless global coverage, full virtualization, and ubiquitous intelligence. These promising features will spawn lots of new verticals and applications, creating an urgent need for on-demand networks. At this point, network slicing has been recognized as an excellent solution to this challenge due to its advantages of flexible customization and logical isolation. Driven by massive data, intelligent network slicing is further endowed with capabilities of real-time perception, accurate prediction, and adaptive decision making, which brings unlimited potential for the development of B5G/6G. To this end, we investigate the application and prospect of intelligent slicing for B5G/6G in three areas, that is, resource allocation, service provisioning, and security. We also list the specific problems in these areas according to each slice lifecycle phase. In addition, a case study of using reinforcement learning for slice admission control is provided to improve the slice admission rate under limited resources and illustrate the advantages of intelligent network slicing.
UR - http://www.scopus.com/inward/record.url?scp=85196671509&partnerID=8YFLogxK
U2 - 10.1109/MWC.021.2200589
DO - 10.1109/MWC.021.2200589
M3 - 文章
AN - SCOPUS:85196671509
SN - 1536-1284
VL - 31
SP - 271
EP - 277
JO - IEEE Wireless Communications
JF - IEEE Wireless Communications
IS - 3
ER -