Exploration of NWDAF Development Architecture for 6G AI-Native Networks

doi:10.12142/ZTECOM.202501006

ZTE Communications ›› 2025, Vol. 23 ›› Issue (1): 45-52.DOI: 10.12142/ZTECOM.202501006

收稿日期:2024-12-15 出版日期:2025-03-25 发布日期:2025-03-25

Exploration of NWDAF Development Architecture for 6G AI-Native Networks

HE Shiwen¹^,²(), PENG Shilin¹, DONG Haolei¹, WANG Liangpeng², AN Zhenyu²

^1.School of Computer Science and Engineering, Central South University, Changsha 410083, China
^2.Purple Mountain Laboratories, Nanjing 210096, China

Received:2024-12-15 Online:2025-03-25 Published:2025-03-25
About author:HE Shiwen ( shiwen.he.hn@csu.edu.cn) is a professor at the School of Computer Science and Engineering, Central South University, China. His research interests include basic theoretical research and standard protocol development in wireless cellular/satellite/WLAN communication and networking, distributed learning and optimization computing, data mining and intelligent analysis, as well as research and development of low-level implementation theory and application technology for open programmable AI-native communication prototype systems.
PENG Shilin received his BS degree in IoT engineering from the School of Internet of Things Engineering, Hohai University, China in 2023. He is currently pursuing his MS degree in computer technology at Central South University, China. His research interest is AI-Native wireless communication.
DONG Haolei received his MS degree in computer science from the School of Computer Science, Wuhan University, China in 2019. He is currently pursuing his PhD degree in computer science at Central South University, China. His research interests include AI-Native wireless communication, 6G core networks, and knowledge graphs.
WANG Liangpeng is a senior engineer at Purple Mountain Laboratories (PML), China, specializing in wireless communication and network technologies. His research focuses on big data analytics and AI algorithms for networks, as well as knowledge graph-driven algorithms for autonomous network operations and intelligence.
AN Zhenyu is currently a senior engineer at Purple Mountain Laboratories (PML), China. His research interests include optimization theory and ultra-reliable and low latency communications.
Supported by:
the National Key Research and Development Program of China(2023YFE0200700);National Natural Science Foundation of China(62171474);ZTE Industry-University-Institute Cooperation Funds(IA20241014013)

摘要/Abstract

Abstract:

Artificial intelligence (AI)-native communication is considered one of the key technologies for the development of 6G mobile communication networks. This paper investigates the architecture for developing the network data analytics function (NWDAF) in 6G AI-native networks. The architecture integrates two key components: data collection and management, and model training and management. It achieves real-time data collection and management, establishing a complete workflow encompassing AI model training, deployment, and intelligent decision-making. The architecture workflow is evaluated through a vertical scaling use case by constructing an AI-native network testbed on Kubernetes. Within this proposed NWDAF, several machine learning (ML) models are trained to make vertical scaling decisions for user plane function (UPF) instances based on data collected from various network functions (NFs). These decisions are executed through the Kubernetes API, which dynamically allocates appropriate resources to UPF instances. The experimental results show that all implemented models demonstrate satisfactory predictive capabilities. Moreover, compared with the threshold-based method in Kubernetes, all models show a significant advantage in response time. This study not only introduces a novel AI-native NWDAF architecture but also demonstrates the potential of AI models to significantly improve network management and resource scaling in 6G networks.

Key words: 6G, AI-native, NWDAF, UPF scaling

. [J]. ZTE Communications, 2025, 23(1): 45-52.

HE Shiwen, PENG Shilin, DONG Haolei, WANG Liangpeng, AN Zhenyu. Exploration of NWDAF Development Architecture for 6G AI-Native Networks[J]. ZTE Communications, 2025, 23(1): 45-52.

图/表 7

Figure 1 System functional architecture

Figure 2 Data collection and management

Figure 3 Model training and management

Figure 4 System deployment structure

Figure 5 Vertical scaling of NF container

Figure 6 Data collection and algorithm comparison during the scaling process

Table 1 Prediction accuracy of the three models

Model	Accuracy/%
RNN	94.74
LSTM	94.87
XGBoost	86.84

参考文献 22

1	YANG B LIANG X, LIU S N, et al. Intelligent 6G wireless network with multi-dimensional information perception [J]. ZTE communications, 2023, 21( 2): 3– 10. DOI: 10.12142/ZTECOM.202302002
2	CHEN Z R, ZHANG Z Y, YANG Z H. Big AI models for 6G wireless networks: opportunities, challenges, and research directions [J]. IEEE wireless communications, 2024, 31( 5): 164– 172. DOI: 10.1109/MWC.015.2300404
3	HE S W, DONG H L, PENG S L, et al. Challenges and methods of constructing a verification system for endogenous intelligent communication in wireless networks [J]. Mobile communications, 2024, 48( 7): 2– 14. DOI: 10.3969/j.issn.1006-1010.20240629-0001
4	GAO Y CHEN J J, LI D P. Intelligence driven wireless networks in B5G and 6G era: a survey [J]. ZTE communications, 2024, 22( 3): 99– 105. DOI: 10.12142/ZTECOM.202403012
5	AGIWAL M, ROY A, SAXENA N. Next generation 5G wireless networks: a comprehensive survey [J]. IEEE communications surveys and tutorials, 2016, 18( 3): 1617– 1655. DOI: 10.1109/COMST.2016.2532458
6	ZAIDI Z, FRIDERIKOS V, YOUSAF Z, et al. Will SDN be part of 5G? [J]. IEEE communications surveys and tutorials, 20( 4): 3220– 3258. DOI: 10.1109/comst.2018.2836315
7	LUONG D H, THIEU H T, OUTTAGARTS A, et al. Cloudification and autoscaling orchestration for container-based mobile networks toward 5G: experimentation, challenges and perspectives [C]//Proc. IEEE 87th Vehicular Technology Conference (VTC Spring). IEEE, 2018: 1– 7. DOI: 10.1109/VTCSpring.2018.8417602
8	QUAN Q. Intelligent and autonomous management in cloud-native future networks: a survey on related standards from an architectural perspective [J]. Future Internet, 2021, 13( 2), 42. DOI: 10.3390/fi13020042
9	HE S W. An endogenous intelligent architecture for wireless communication networks [J]. Wireless networks, 2024, 30( 2): 1069– 1084. DOI: 10.1007/s11276-023-03545-9
10	3GPP. Technical specification group services and system aspects; achitecture enhancements for 5G system (5GS) to support network data analytics services: TS 23.288 [S]. 2023
11	LEE J, SOLAT F, KIM T Y, et al. Federated learning-empowered mobile network management for 5G and beyond networks: from access to core [J]. IEEE communications surveys and tutorials, 2024, 26( 3): 2176– 2212. DOI: 10.1109/COMST.2024.3352910
12	SAAD W, BENNIS M, CHEN M Z. A vision of 6G wireless systems: applications, trends, technologies, and open research problems [J]. IEEE network, 2020, 34( 3): 134– 142. DOI: 10.1109/MNET.001.1900287
13	GKONIS P K, NOMIKOS N, TRAKADAS P, et al. Leveraging network data analytics function and machine learning for data collection, resource optimization, security and privacy in 6G networks [J]. IEEE access, 2024, 12: 21320– 21336. DOI: 10.1109/ACCESS.2024.3359992
14	WU W, ZHOU C H, LI M S, et al. AI-native network slicing for 6G networks [J]. IEEE wireless communications, 29( 1): 96– 103. DOI: 10.1109/mwc.001.2100338
15	MEKRACHE A, BOUTIBA K, KSENTINI A. Combining network data analytics function and machine learning for abnormal traffic detection in beyond 5G [C]//Proc. IEEE Global Communications Conference. IEEE, 2023: 1204– 1209. DOI: 10.1109/GLOBECOM54140.2023.10436766
16	NISHA L K, KUMAR R. A smart data analytics system generating for 5G N/W system via ML based algorithms for the better communications [C]// Proc. 1st International Conference on Innovative Sustainable Technologies for Energy, Mechatronics, and Smart Systems (ISTEMS). IEEE, 2024: 1– 6. DOI: 10.1109/ISTEMS60181.2024.10560068
17	SEVGICAN S, TURAN M, GÖKARSLAN K, et al. Intelligent network data analytics function in 5G cellular networks using machine learning [J]. Journal of communications and networks, 2020, 22( 3): 269– 280. DOI: 10.1109/JCN.2020.000019
18	MANIAS D M, CHOUMAN A, SHAMI A. An NWDAF approach to 5G core network signaling traffic: analysis and characterization [C]//Proc. IEEE Global Communications Conference. IEEE, 2022: 6001– 6006. DOI: 10.1109/GLOBECOM48099.2022.10000989
19	ZHANG C J, SHAN G Y, ROH B H. Fair federated learning for multi-task 6G NWDAF network anomaly detection [EB/OL]. ( 2024-09-25)[ 2024-10-09].
20	SANTOS G L, ENDO P T, SADOK D, et al. When 5G meets deep learning: a systematic review [J]. Algorithms, 2020, 13( 9): 208. DOI: 10.3390/a13090208
21	TEZERGIL B, ONUR E. Wireless backhaul in 5G and beyond: issues, challenges and opportunities [J]. IEEE communications surveys and tutorials, 2022, 24( 4): 2579– 2632. DOI: 10.1109/COMST.2022.3203578
22	LY A, YAO Y D. A review of deep learning in 5G research: channel coding, massive MIMO, multiple access, resource allocation, and network security [J]. IEEE open journal of the communications society, 2021, 2: 396– 408. DOI: 10.1109/OJCOMS.2021.3058353

编辑推荐 0

Metrics

阅读次数

全文

128

HTML			PDF

最新录用	在线预览	正式出版	最新录用	在线预览	正式出版
0	0	5	0	0	123

来源	本网站	其他网站

次数	98	30
比例	77%	23%

摘要

122

最新录用	在线预览	正式出版

0	0	122

	来源	本网站

	次数	122
	比例	100%

Exploration of NWDAF Development Architecture for 6G AI-Native Networks

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 7

参考文献 22

相关文章 0

编辑推荐 0

Metrics