An architecture and performance evaluation framework for artificial intelligence solutions in beyond 5G radio access networks

EURASIP Journal on Wireless Communications and Networking

Research

Qing He¹  György Dán¹  Georgios P. Koudouridis² 

[1] Network and Systems Engineering, KTH Royal Institute of Technology, Stockholm, Sweden;Wireless Systems Laboratory, Stockholm Research Centre, Huawei Technologies Sweden, Kista, Sweden;

关键词: Artificial intelligence;    Machine learning;    Radio access networks;    Network automation;   

DOI  :  10.1186/s13638-022-02164-w

 received in 2022-05-17, accepted in 2022-08-25,  发布年份 2022

来源: Springer

【 摘 要 】

The evolution of mobile communications towards beyond 5th-generation (B5G) networks is envisaged to incorporate high levels of network automation. Network automation requires the development of a network architecture that accommodates multiple solutions based on artificial intelligence (AI) and machine learning (ML). Consequently, integrating AI into the 5th-generation (5G) systems such that we could leverage the advantages of ML techniques to optimize and improve the networks is one challenging topic for B5G networks. Based on a review of 5G system architecture, the state-of-the-art candidate AI/ML techniques, and the progress of the state of the art, and the on AI/ML for 5G in standards we define an AI architecture and performance evaluation framework for the deployment of the AI/ML solution in B5G networks. The suggested framework proposes three AI architectures alternatives, a centralized, a completely decentralized and an hybrid AI architecture. More specifically, the framework identifies the logical AI functions, determines their mapping to the B5G radio access network architecture and analyses the associated deployment cost factors in terms of compute, communicate and store costs. The framework is evaluated based on a use case scenario for heterogeneous networks where it is shown that the deployment cost profiling is different for the different AI architecture alternatives, and that this cost should be considered for the deployment and selection of the AI/ML solution.

【 授权许可】

CC BY   
© The Author(s) 2022

【 图 表 】

41116_2022_35_Article_IEq517.gif

41116_2022_35_Article_IEq516.gif

41116_2022_35_Article_IEq510.gif

41116_2022_35_Article_IEq509.gif

Fig. 28

41116_2022_35_Article_IEq507.gif

41116_2022_35_Article_IEq506.gif

Fig. 27

41116_2022_35_Article_IEq504.gif

41116_2022_35_Article_IEq503.gif

Fig. 26

41116_2022_35_Article_IEq501.gif

41116_2022_35_Article_IEq500.gif

41116_2022_35_Article_IEq499.gif

41116_2022_35_Article_IEq498.gif

41116_2022_35_Article_IEq497.gif

41116_2022_35_Article_IEq496.gif

41116_2022_35_Article_IEq495.gif

41116_2022_35_Article_IEq494.gif

41116_2022_35_Article_IEq493.gif

41116_2022_35_Article_IEq492.gif

41116_2022_35_Article_IEq491.gif

Fig. 25

Fig. 24

41116_2022_35_Article_IEq489.gif

Fig. 23

41116_2022_35_Article_IEq487.gif

Fig. 22

41116_2022_35_Article_IEq485.gif

Fig. 21

Fig. 5

41116_2022_35_Article_IEq481.gif

Fig. 4

Fig. 3

41116_2022_35_Article_IEq478.gif

41116_2022_35_Article_IEq476.gif

Fig. 1

41116_2022_35_Article_IEq474.gif

41116_2022_35_Article_IEq473.gif

41116_2022_35_Article_IEq472.gif

Fig. 18

41116_2022_35_Article_IEq470.gif

41116_2022_35_Article_IEq469.gif

41116_2022_35_Article_IEq468.gif

41116_2022_35_Article_IEq467.gif

41116_2022_35_Article_IEq466.gif

41116_2022_35_Article_IEq465.gif

41116_2022_35_Article_IEq464.gif

Fig. 2

41116_2022_35_Article_IEq462.gif

Fig. 1

41116_2022_35_Article_IEq460.gif

41116_2022_35_Article_IEq459.gif

41116_2022_35_Article_IEq458.gif

41116_2022_35_Article_IEq457.gif

41116_2022_35_Article_IEq456.gif

Fig. 17

41116_2022_35_Article_IEq454.gif

Fig. 16

41116_2022_35_Article_IEq452.gif

Fig. 15

41116_2022_35_Article_IEq450.gif

Fig. 14

41116_2022_35_Article_IEq448.gif

Fig. 13

41116_2022_35_Article_IEq446.gif

Fig. 12

41116_2022_35_Article_IEq444.gif

Fig. 2

Fig. 11

41116_2022_35_Article_IEq441.gif

41116_2022_35_Article_IEq440.gif

41116_2022_35_Article_IEq439.gif

Fig. 1

Fig. 7

41116_2022_35_Article_IEq425.gif

12888_2022_4443_Article_IEq7.gif

12888_2022_4443_Article_IEq6.gif

12888_2022_4443_Article_IEq5.gif

41116_2022_35_Article_IEq421.gif

41116_2022_35_Article_IEq420.gif

41116_2022_35_Article_IEq419.gif

Fig. 6

41116_2022_35_Article_IEq417.gif

41116_2022_35_Article_IEq416.gif

41116_2022_35_Article_IEq414.gif

41116_2022_35_Article_IEq413.gif

41116_2022_35_Article_IEq412.gif

41116_2022_35_Article_IEq411.gif

41116_2022_35_Article_IEq410.gif

41116_2022_35_Article_IEq409.gif

41116_2022_35_Article_IEq408.gif

Fig. 4

41116_2022_35_Article_IEq406.gif

41116_2022_35_Article_IEq405.gif

41116_2022_35_Article_IEq404.gif

41116_2022_35_Article_IEq403.gif

12888_2022_4443_Article_IEq4.gif

12888_2022_4443_Article_IEq3.gif

12888_2022_4443_Article_IEq2.gif

12888_2022_4443_Article_IEq1.gif

41116_2022_35_Article_IEq398.gif

41116_2022_35_Article_IEq397.gif

41116_2022_35_Article_IEq396.gif

41116_2022_35_Article_IEq395.gif

41116_2022_35_Article_IEq394.gif

41116_2022_35_Article_IEq393.gif

Fig. 3

Fig. 4

Fig. 3

41116_2022_35_Article_IEq387.gif

41116_2022_35_Article_IEq386.gif

Fig. 2

Fig. 4

41116_2022_35_Article_IEq383.gif

41116_2022_35_Article_IEq382.gif

41116_2022_35_Article_IEq381.gif

41116_2022_35_Article_IEq380.gif

41116_2022_35_Article_IEq379.gif

Fig. 3

41116_2022_35_Article_IEq377.gif

Fig. 2

41116_2022_35_Article_IEq375.gif

Fig. 1

41116_2022_35_Article_IEq373.gif

41116_2022_35_Article_IEq372.gif

Fig. 3

Fig. 2

41116_2022_35_Article_IEq369.gif

Fig. 1

41116_2022_35_Article_IEq367.gif

41116_2022_35_Article_IEq366.gif

Fig. 1

41116_2022_35_Article_IEq364.gif

41116_2022_35_Article_IEq363.gif

41116_2022_35_Article_IEq362.gif

41116_2022_35_Article_IEq361.gif

41116_2022_35_Article_IEq360.gif

41116_2022_35_Article_IEq359.gif

41116_2022_35_Article_IEq350.gif

41116_2022_35_Article_IEq349.gif

41116_2022_35_Article_IEq348.gif

41116_2022_35_Article_IEq347.gif

41116_2022_35_Article_IEq346.gif

41116_2022_35_Article_IEq345.gif

41116_2022_35_Article_IEq344.gif

41116_2022_35_Article_IEq343.gif

41116_2022_35_Article_IEq342.gif

41116_2022_35_Article_IEq341.gif

41116_2022_35_Article_IEq340.gif

41116_2022_35_Article_IEq339.gif

41116_2022_35_Article_IEq338.gif

41116_2022_35_Article_IEq337.gif

41116_2022_35_Article_IEq336.gif

41116_2022_35_Article_IEq335.gif

41116_2022_35_Article_IEq334.gif

41116_2022_35_Article_IEq333.gif

Fig. 2

Fig. 2

41116_2022_35_Article_IEq330.gif

41116_2022_35_Article_IEq321.gif

41116_2022_35_Article_IEq318.gif

41116_2022_35_Article_IEq304.gif

41116_2022_35_Article_IEq300.gif

Fig. 2

41116_2022_35_Article_IEq286.gif

41116_2022_35_Article_IEq283.gif

41116_2022_35_Article_IEq279.gif