Toward AI-Agent-Native 6G Networks: A Survey on Protocols, Multimodal Coordination, and ISCC-Driven Dynamic Networking

doi:10.12142/ZTECOM.202602003

ZTE Communications ›› 2026, Vol. 24 ›› Issue (2): 16-25.DOI: 10.12142/ZTECOM.202602003

• Special Topic • Previous Articles Next Articles

Toward AI-Agent-Native 6G Networks: A Survey on Protocols, Multimodal Coordination, and ISCC-Driven Dynamic Networking

Zhang Xiaotian¹, Xiao Han¹(), Wang Dan², Huang Zhenglei², Xu Changqiao¹

^1.Beijing University of Posts and Telecommunications, Beijing 100876, China
^2.China Mobile Communications Group Co. , Ltd. , Beijing 100032, China

Received:2025-02-03 Online:2026-06-16 Published:2026-06-16
About author:Zhang Xiaotian received his BS degree in the School of Computer Science from Beijing University of Posts and Telecommunications (BUPT), China in 2024. He is currently working toward a doctoral degree in the School of Computer Science, BUPT. His research interests include multi-agent communication and AI.
Xiao Han (xiaohan@bupt.edu.cn) is currently an associate professor with the State Key Laboratory of Networking and Switching Technology, Beijing University of Posts and Telecommunications, China. His research interests include immersive media transmission, online learning, AIGC video services, and generative AI.
Wang Dan is a project manager with the China Mobile Research Institute, China Mobile Communications Group Co., Ltd. She has been actively engaged in standardization research on AI-native and agent-oriented communication networks and has served as a principal contributor to the SA2 NET4AI direction. He work focuses on network architecture evolution, intelligent agent communication, and standardization pathways for integrating AI agents into future mobile networks.
Huang Zhenglei is a director, researcher, and senior engineer with the China Mobile Research Institute, China Mobile Communications Group Co., Ltd., China. He serves as the leader of the Agent Communication Network Working Subgroup under CCSA TC5 WG12. His research interests include future network architecture, intelligent communication networks, agent communication networks, and standardization for AI-native mobile communication systems.
Xu Changqiao received his PhD degree from the Institute of Software, Chinese Academy of Sciences (ISCAS) in January 2009. He was an assistant research fellow and research and development project manager at ISCAS from 2002 to 2007. He was a researcher at Athlone Institute of Technology, Ireland, and a jointly trained PhD candidate at Dublin City University, Ireland, from 2007 to 2009. He joined Beijing University of Posts and Telecommunications (BUPT), China in 2009. He is currently a professor with the State Key Laboratory of Networking and Switching Technology and the Director of the Network Architecture Research Center, BUPT. His research interests include network security, mobile networking, multimedia communications, and future Internet technology. He has published over 200 technical papers in prestigious international journals and conferences, including IEEE Communications Magazine, IEEE/ACM Transactions on Networking, IEEE Transactions on Mobile Computing, INFOCOM, and ACM Multimedia.
Supported by:
the National Science and Technology Major Project of China on Mobile Information Networks(2025ZD1304700);the National Natural Science Foundation of China (NSFC)(62301070);the National Natural Science Foundation of China (NSFC)(62225105);the National Natural Science Foundation of China (NSFC)(62394323);It is also funded by the Beijing University of Posts and Telecommunications?China Mobile Communications Group Co., Ltd. Joint Institute, the Research Initiation Project for Introduced Talents of BUPT(2025KYQD12);the Foundation of the State Key Laboratory of Networking and Switching Technology, Beijing University of Posts and Telecommunications(NST20250303)

Abstract

Abstract:

As 6G approaches, the proliferation of large language models (LLMs) and embodied intelligence is driving a paradigm shift from the Internet of Things (IoT) to the Internet of Agents (IoA). However, traditional network architectures, designed for content-agnostic data transmission, struggle to accommodate the bursty, reasoning-driven traffic patterns and rigorous multimodal synchronization requirements of autonomous agents. This paper surveys the AI-agent communication network (ACN), aiming to bridge the gap between static network resources and dynamic agent tasks. We analyze the evolution from bit-oriented transmission to agentic syntax protocols, which enable intent-based signaling and semantic compression. Furthermore, we explore mechanisms for multi-agent collaborative consensus and distributed decision-making under the constraints of unstable wireless environments. We critically focus on task-driven dynamic networking, examining how integrated sensing, communication, and computing (ISCC) and network-embedded agents (NEA) facilitate the real-time generation of task graphs and intent-aware traffic scheduling. To synthesize these technologies, we propose a reference framework, the Deep-Agentic Network Architecture (DAN-Arch), which vertically integrates physical-layer sensing with application-layer reasoning flows. Finally, open challenges regarding energy efficiency, cross-domain governance, and 3GPP standardization pathways are discussed to guide future research towards a fully agent-native 6G ecosystem.

Key words: AI-agent communication network (ACN), 6G architecture, network-embedded agent (NEA), integrated sensing, communication, and computing (ISCC), multi-agent coordination

Zhang Xiaotian, Xiao Han, Wang Dan, Huang Zhenglei, Xu Changqiao. Toward AI-Agent-Native 6G Networks: A Survey on Protocols, Multimodal Coordination, and ISCC-Driven Dynamic Networking[J]. ZTE Communications, 2026, 24(2): 16-25.

Figures/Tables 9

References 26

[1]	Letaief K B, Chen W, Shi Y M, et al. The roadmap to 6G: AI empowered wireless networks [J]. IEEE communications magazine, 2019, 57(8): 84–90. DOI: 10.1109/mcom.2019.1900271
[2]	Chen Z Q, Sun Q, Li N, et al. Enabling mobile AI agent in 6G era: architecture and key technologies [J]. IEEE network, 2024, 38(5): 66–75. DOI: 10.1109/mnet.2024.3422309
[3]	Wang C X, You X H, Gao X Q, et al. On the road to 6G: visions, requirements, key technologies, and testbeds [J]. IEEE communications surveys & tutorials, 2023, 25(2): 905–974. DOI: 10.1109/comst.2023.3249835
[4]	Zhang Z Q, Xiao Y, Ma Z, et al. 6G wireless networks: vision, requirements, architecture, and key technologies [J]. IEEE vehicular technology magazine, 2019, 14(3): 28–41. DOI: 10.1109/mvt.2019.2921208
[5]	Wei J, Wang X, Schuurmans D, et al. Chain-of-thought prompting elicits reasoning in large language models [PP/OL]. arXiv (2023-01-10) [2026-01-12].
[6]	Nguyen D C, Ding M, Pathirana P N, et al. 6G Internet of Things: a comprehensive survey [J]. IEEE Internet of Things journal, 2022, 9(1): 359–383. DOI: 10.1109/jiot.2021.3103320
[7]	Chen M K, Liu M H, Wang C Y, et al. Cross-modal graph semantic communication assisted by generative AI in the metaverse for 6G [J]. Research, 2024, 7: 342. DOI: 10.34133/research.0342
[8]	Wang Z R, Chen M K, Liu Q. A review on multimodal communications for human-robot collaboration in 5G: from visual to tactile [J]. Intelligence & robotics, 2025, 5(3): 579–606. DOI: 10.20517/ir.2025.30
[9]	Wang R J, Chen K M, Li S J, et al. Efficient semantic codec for real-time vibrotactile transmission [C]//The 33rd ACM International Conference on Multimedia. ACM, 2025: 12092–12101. DOI: 10.1145/3746027.3755630
[10]	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
[11]	Qin Z, Tao X, Lu J, et al. Semantic communications: principles and challenges [PP/OL]. arXiv (2022-06-27) [2026-01-12].
[12]	Chen J, Zhang H, Xie Z. Space-air-ground integrated network (SAGIN): a survey [PP/OL]. arXiv (2023-07-27) [2026-01-12].
[13]	Shao J W, Mao Y Y, Zhang J. Task-oriented communication for multidevice cooperative edge inference [J]. IEEE transactions on wireless communications, 2023, 22(1): 73–87. DOI: 10.1109/twc.2022.3191118
[14]	Zhang P, Xu W J, Gao H, et al. Toward wisdom-evolutionary and primitive-concise 6G: a new paradigm of semantic communication networks [J]. Engineering, 2022, 8: 60–73. DOI: 10.1016/j.eng.2021.11.003
[15]	Strinati E C, Barbarossa S. 6G Networks: beyond Shannon towards semantic and goal-oriented communications [PP/OL]. arXiv (2021-02-17) [2026-01-12].
[16]	Liang C, Du H, Sun Y, et al. Generative AI-driven semantic communication networks: architecture, technologies and applications [PP/OL]. arXiv(2024-01-07) [2026-01-12].
[17]	Dong R, She C Y, Hardjawana W, et al. Deep learning for hybrid 5G services in mobile edge computing systems: learn from a digital twin [J]. IEEE transactions on wireless communications, 2019, 18(10): 4692–4707. DOI: 10.1109/twc.2019.2927312
[18]	Liu F, Zhou L F, Masouros C, et al. Toward dual-functional radar-communication systems: optimal waveform design [J]. IEEE transactions on signal processing, 2018, 66(16): 4264–4279. DOI: 10.1109/tsp.2018.2847648
[19]	Yang Z, Chen M, Wong K-K, et al. Federated learning for 6G: applications, challenges, and opportunities [PP/OL]. arXiv (2021-01-05) [2026-01-12].
[20]	Xie H Q, Qin Z J, Li G Y, et al. Deep learning enabled semantic communication systems [J]. IEEE transactions on signal processing, 2021, 69: 2663–2675. DOI: 10.1109/tsp.2021.3071210
[21]	Vetriveeran D, Leena Sri R. Resource provisioning in fog computing—a survey [J]. ACM computing surveys, 2025, 57(12): 1–26. DOI: 10.1145/3744662
[22]	Chen M, Hao Y X, Hwang K, et al. Disease prediction by machine learning over big data from healthcare communities [J]. IEEE access, 2017, 5: 8869–8879. DOI: 10.1109/access.2017.2694446
[23]	Pang L, Yang C G, Chen D Y, et al. A survey on intent-driven networks [J]. IEEE access, 2020, 8: 22862–22873. DOI: 10.1109/access.2020.2969208
[24]	Uysal E, Kaya O, Baghaee S, et al. Age of information in practice [PP/OL]. arXiv (2021-06-02) [2026-01-12].
[25]	Akyildiz I F, Han C, Nie S. Combating the distance problem in the millimeter wave and terahertz frequency bands [J]. IEEE communications magazine, 2018, 56(6): 102–108. DOI: 10.1109/mcom.2018.1700928
[26]	Mao Y Y, You C S, Zhang J, et al. A survey on mobile edge computing: the communication perspective [J]. IEEE communications surveys & tutorials, 2017, 19(4): 2322–2358. DOI: 10.1109/comst.2017.2745201

Architecture Plane	Prior Work Focus (semantic communication and 5G)	Proposed 6G-DAN-Arch (Agent-Native)
Collective intelligence	Generative 3D & vibrotactile codecs^{[7, 9]}	Agentic syntax & distributed consensus
ISCC control	Reactive multimodal QoS^[8]	Predictive ISCC task-triggered slicing
Agentic resource	Edge-cloud offloading	NEA

Architecture Plane	Prior Work Focus (semantic communication and 5G)	Proposed 6G-DAN-Arch (Agent-Native)
Collective intelligence	Generative 3D & vibrotactile codecs^{[7, 9]}	Agentic syntax & distributed consensus
ISCC control	Reactive multimodal QoS^[8]	Predictive ISCC task-triggered slicing
Agentic resource	Edge-cloud offloading	NEA

Traffic Category	Typical Application Scenario	Temporal Pattern	Payload Type	Key Network Requirement
Legacy mobile broadband	4K/8K video streaming, web browsing	Continuous or periodic	High-redundancy bitstreams	Throughput (Mbit/s)
xURLLC (Embodied)	Industrial robots, haptic feedback	Isochronous, high-frequency	Short control packets	Deterministic latency (<0.1 ms), jitter control
Generative AI streams	Large language models, CoT reasoning	Non-periodic, bursty (token spikes)	Semantic vectors, knowledge graphs	Semantic fidelity, burst tolerance
NTN swarm telemetry	UAV swarms, satellite IoT	Intermittent, topology-dependent	State updates, sparse data	Routing convergence, topology robustness

Traffic Category	Typical Application Scenario	Temporal Pattern	Payload Type	Key Network Requirement
Legacy mobile broadband	4K/8K video streaming, web browsing	Continuous or periodic	High-redundancy bitstreams	Throughput (Mbit/s)
xURLLC (Embodied)	Industrial robots, haptic feedback	Isochronous, high-frequency	Short control packets	Deterministic latency (<0.1 ms), jitter control
Generative AI streams	Large language models, CoT reasoning	Non-periodic, bursty (token spikes)	Semantic vectors, knowledge graphs	Semantic fidelity, burst tolerance
NTN swarm telemetry	UAV swarms, satellite IoT	Intermittent, topology-dependent	State updates, sparse data	Routing convergence, topology robustness

Feature Dimension	Traditional 5G/AI4Net	Proposed 6G/Agent4Net
Primary user	Humans and passive IoT sensors	Autonomous AI agents & embodied robots
Core function	Transparent data pipe (content-agnostic)	Intelligent substrate (intent-aware)
Protocol stack	TCP/IP + standard NAS	Agentic syntax and new NAS (intent-signaling)
Resource scheduling	Reactive (based on channel quality/CQI)	Predictive (based on task graph & ISCC)
Network topology	Static/semi-static slicing	Dynamic (task-triggered ephemeral subnets)
Routing mechanism	IP address (shortest path)	A-DNS (capability-based routing)
Role of AI	AI optimizes the network (overlay)	Network is built for AI (native design)

Toward AI-Agent-Native 6G Networks: A Survey on Protocols, Multimodal Coordination, and ISCC-Driven Dynamic Networking

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 9

References 26

Related Articles 0

Recommended Articles

Metrics