Intent-Driven Control System for Heterogeneous Agent-Oriented Networking (HaoNet)

doi:10.12142/ZTECOM.202602007

ZTE Communications ›› 2026, Vol. 24 ›› Issue (2): 52-63.DOI: 10.12142/ZTECOM.202602007

• Special Topic • Previous Articles Next Articles

Intent-Driven Control System for Heterogeneous Agent-Oriented Networking (HaoNet)

Wang Bowen¹, Lu Lu², Li Huimin¹, Yang Chungang¹()

^1.Xidian University, Xi’an 710071, China
^2.China Mobile Research Institute, Beijing 100053, China

Received:2026-03-11 Online:2026-06-16 Published:2026-06-16
About author:Wang Bowen is currently pursuing his doctoral degree with the GUIDE family at Xidian University, China. His research interests include intent-driven networks, large AI models, and software-defined networks.
Lu Lu received her master’s degree from Beijing University of Posts and Telecommunications, China. She is currently the Deputy Director of the Department of Basic Network Technology, China Mobile Research Institute; the leader of the Core Network Group, CCSA TC5; and the Vice Chair of ITU-T SG13. Her main research interests include 5G/6G network architecture and next-generation IP technology.
Li Huimin is currently pursuing a master’s degree in the GUIDE family at Xidian University, China. Her research interests include intent-driven networks, agents, and software-defined networks.
Yang Chungang (guideyang2050@163.com) is a full professor at Xidian University, China. His research interests are AI-enabled 6G wireless mobile networks, intent-driven networks, space-terrestrial networks, and game theory for emerging communication networks.

Abstract

Abstract:

Empowered by advances in large language models, the growing integration of autonomous agents into industrial and daily-life sectors is turning them into new networking entities. Such agent-oriented networking features high interaction frequencies and emergent task-driven structures, necessitating strong network policy consistency and reliability within dynamic environments. To address these challenges, we propose a network control system that integrates Intent-Driven Network (IDN) into Heterogeneous Agent-Oriented Networking (HaoNet). IDN focuses on high-level task intents and provides flexible reconfiguration and adaptive optimization, thereby enhancing the effectiveness of agent-oriented networking. In this paper, we first summarize three key features of HaoNet: task-driven operation, distributed collaboration, and closed-loop intelligence. Furthermore, we propose a comprehensive system architecture, which includes the application layer, the intent layer, and the infrastructure layer, and investigate the associated key technologies. Finally, typical application scenarios are presented to demonstrate the practical value of the proposed system in enabling robust agent-oriented networking control.

Key words: AI agent, agent-oriented networking, intent-driven network, network control

Wang Bowen, Lu Lu, Li Huimin, Yang Chungang. Intent-Driven Control System for Heterogeneous Agent-Oriented Networking (HaoNet)[J]. ZTE Communications, 2026, 24(2): 52-63.

Figures/Tables 10

Figure 1 A general networking scenario of collaborative infrastructure maintenance in HaoNet

Figure 2 Three task orchestration modes in HaoNet

Figure 3 General internal architecture of agent

Figure 4 Architecture of an intent-driven control system for HaoNet

Figure 5 Internal MAPE-K architecture of the agent

Figure 6 Five control patterns of MAPE-K across three task orchestration modes

Figure 7 Technical architecture of LLM-based intent recognition, refinement, decomposition, and mapping

Figure 8 Schematic of CoX mechanism

Figure 9 Schematic of large-small model collaboration mechanism

Figure 10 Two application scenarios of HaoNet: smart home and smart factory

References 34

[1]	ETSI GR ENI 051 v4.1.1: 2025 Experiential networked intelligence (ENI); study on AI agents based next-generation network slicing [S]
[2]	Monica. Manus [EB/OL]. [2026-03-10].
[3]	Bytedance. Coze [EB/OL]. [2026-03-10].
[4]	China Mobile Think Tank. AI-agent communication network white paper [R]. Beijing: China Mobile, 2024
[5]	Yang C G. Intent-driven autonomous network [M]. Beijing: Science Press, 2024
[6]	Gupta M, Acharya V. Agnet: a novel AI agent network architecture [PP/OL]. SSRN [2026-03-10].
[7]	Wang Y T, Guo S L, Pan Y H, et al. Internet of agents: fundamentals, applications, and challenges [J]. IEEE transactions on cognitive communications and networking, 2026, 12: 4476–4501. DOI: 10.1109/TCCN.2025.3623369
[8]	Anthropic. Model context protocol (MCP)[EB/OL]. [2026-03-10].
[9]	Google. Agent to agent protocol (A2A) [EB/OL]. [2026-03-10].
[10]	Xiao Y, Shi G M, Zhang P. Toward agentic AI networking in 6G: a generative foundation model-as-agent approach [J]. IEEE communications magazine, 2025, 63(9): 68–74. DOI: 10.1109/MCOM.001.2500005
[11]	Du C G, Wang C Y, Chao Y H, et al. AI agent communication from Internet architecture perspective: challenges and opportunities [PP/OL]. V1. arXiv (2025-09-02)[2026-03-10].
[12]	Minhas S, Jaswal R, Sharma A, et al. Revolutionizing networking: a comprehensive overview of intent-based networking [C]//Proceedings of International Conference on Emerging Innovations and Advanced Computing (INNOCOMP). IEEE, 2024: 463–468. DOI: 10.1109/INNOCOMP63224.2024.00081
[13]	Jiang G Z, Wang K Z, Chen X M, et al. Agentic AI empowered intent-based networking for 6G [PP/OL]. V1. arXiv (2026-01-10)[2026-03-10].
[14]	Lee K Y. Intent-driven autonomous reconfiguration for telecom OSS: a generative-AI-empowered hierarchical multi-agent architecture [C]//Proceedings of 25th Asia-Pacific Network Operations and Management Symposium (APNOMS). IEEE, 2025: 1–6. DOI: 10.23919/APNOMS67058.2025.11181316
[15]	Romero M L, Suyama R. Agentic AI for intent-based industrial automation [C]//Proceedings of 16th IEEE International Conference on Industry Applications (INDUSCON). IEEE, 2025: 437–444. DOI: 10.1109/INDUSCON66435.2025.11241317
[16]	Chen S T, Liao Q, Aijaz A, et al. Goal-oriented multi-agent semantic networking: unifying intents, semantics, and intelligence [PP/OL]. V2. arXiv (2026-03-24)[2026-03-10].
[17]	Tong J W, Guo W, Shao J W, et al. WirelessAgent: large language model agents for intelligent wireless networks [PP/OL]. V1. arXiv (2025-05-02)[2026-03-10].
[18]	Dzeparoska K, Beigi-Mohammadi N, Tizghadam A, et al. Towards a self-driving management system for the automated realization of intents [J]. IEEE access, 2021, 9: 159882–159907. DOI: 10.1109/ACCESS.2021.3129990
[19]	Weyns D, Schmerl B, Grassi V, et al. On patterns for decentralized control in self-adaptive systems [M]//Software Engineering for Self-Adaptive Systems II. Berlin, Heidelberg: Springer, 2013: 76–107. DOI: 10.1007/978-3-642-35813-5_4
[20]	Jiang F B, Peng Y B, Dong L, et al. Large language model enhanced multi-agent systems for 6G communications [J]. IEEE wireless communications, 2024, 31(6): 48–55. DOI: 10.1109/MWC.016.2300600
[21]	Saad W, Hashash O, Thomas C K, et al. Artificial general intelligence (AGI)-native wireless systems: a journey beyond 6G [J]. Proceedings of the IEEE, 2025, 113(9): 849–887. DOI: 10.1109/JPROC.2025.3526887
[22]	Cui H W, Du Y Y, Yang Q, et al. LLMind: orchestrating AI and IoT with LLM for complex task execution [J]. IEEE communications magazine, 2025, 63(4): 214–220. DOI: 10.1109/MCOM.002.2400106
[23]	Chen Y X, Li R P, Zhao Z F, et al. NetGPT: a native-AI network architecture beyond provisioning personalized generative services [PP/OL]. V4. arXiv (2024-03-09)[2026-03-10].
[24]	Zou H, Zhao Q Y, Tian Y, et al. TelecomGPT: a framework to build telecom-specific large language models [J]. IEEE transactions on machine learning in communications and networking, 2025, 3: 948–975. DOI: 10.1109/TMLCN.2025.3593184
[25]	Wang X C, Ji S W, Li C. Research on intelligent computing network technology for large-scale pre-trained models [J]. Telecommunications science, 2024, 40(6): 160–172. DOI: 10.11959/j.issn.1000-0801.2024167
[26]	Huang B, Wu S A, Wang W G, et al. KG-LLM-MCom: a survey on integration of knowledge graph and large language model [J]. Journal of Wuhan University (natural science edition), 2024, 70(4): 397–412. DOI: 10.14188/j.1671-8836.2024.0040
[27]	Hu Z Q, Wang L, Lan Y H, et al. LLM-adapters: an adapter family for parameter-efficient fine-tuning of large language models [C]//Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing. Stroudsburg, PA, USA: ACL, 2023: 5254–5276. DOI: 10.18653/v1/2023.emnlp-main.319
[28]	Mekrache A, Ksentini A. LLM-enabled intent-driven service configuration for next generation networks [C]//Proceedings of IEEE 10th International Conference on Network Softwarization (NetSoft). IEEE, 2024: 253–257. DOI: 10.1109/NetSoft60951.2024.10588881
[29]	Lester B, Al-Rfou R, Constant N. The power of scale for parameter-efficient prompt tuning [C]//Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing. ACL, 2021: 3045–3059. DOI: 10.18653/v1/2021.emnlp-main.243
[30]	Hu E J, Shen Y, Wallis P, et al. LoRA: low-rank adaptation of large language models [C]//International Conference on Learning Representations. ICLR, 2022: 1–20. DOI: 10.48550/arXiv.2106.09685
[31]	Han H Y, Wang Y, Shomer H, et al. Retrieval-augmented generation with graphs (GraphRAG) [PP/OL]. V2. arXiv (2025-01-08) [2026-03-10].
[32]	Edge D, Trinh H, Cheng N, et al. From local to global: a graph RAG approach to query-focused summarization [PP/OL]. V2. arXiv (2025-02-19)[2026-03-10].
[33]	Ou Y Y. Intelligent intent policy refinement of space-terrestrial network [D]. Xi‘an: Xidian University, 2023
[34]	Liu Y Q, Liu G Y, Wang J C, et al. LAMeTA: intent-aware agentic network optimization via a large AI model-empowered two-stage approach [J]. IEEE journal on selected areas in communications, 2026, 44: 2462–2478. DOI: 10.1109/JSAC.2025.3642840

Intent-Driven Control System for Heterogeneous Agent-Oriented Networking (HaoNet)

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 10

References 34

Related Articles 1

Recommended Articles

Metrics