Similar articles

Abstract: The booming of artificial intelligence (AI) agents has brought about promising business scenarios for sixth-generation (6G) mobile networks, while simultaneously posing significant challenges to network functionalities and infrastructure. These AI agents can be deployed on end devices (e.g., intelligent robots and intelligent cars) or as digital entities (e.g., personal AI assistants). As novel service entities with autonomous decision-making and task execution capabilities, AI agents introduce potential risks of uncontrollable actions and privacy disclosures. AI agents also require new 6G capabilities beyond traditional communication, including multimodality information interaction (e.g., AI models and tokens) and support for service requirements (e.g., computing and sensing of data). In this article, we introduce the concept of AI-agent communication network (ACN), a new paradigm to enable global information interaction and on-demand capability provisioning for single or multiple AI agents. We first introduce the vision and architectural framework of ACN. Then, key technologies and future research directions related to ACN are discussed. Furthermore, we provide potential use cases to elaborate on how ACN can expand the service capabilities of 6G networks.

面向6G的智能体通信网络：愿景、架构与关键技术

[1]Adornetto C, Mora A, Hu K, et al., 2025. Generative agents in agent-based modeling: overview, validation, and emerging challenges. IEEE Trans Artif Intell, 6(12):3165-3183.

[2]Bai CJ, Xu HZ, Li XL, 2024. Embodied-AI with large models: research and challenges. Sci Sin Inform, 54(9):2035-2082 (in Chinese).

[3]Bariah L, Zhao QY, Zou H, et al., 2024. Large generative AI models for telecom: the next big thing? IEEE Commun Mag, 62(11):84-90.

[4]Brohan A, Brown N, Carbajal J, et al., 2023. RT-1: robotics Transformer for real-world control at scale.

[5]Chai HY, Wang HD, Li T, et al., 2024. Generative AI-driven digital twin for mobile networks. IEEE Netw, 38(5):84-92.

[6]Chen ZQ, Sun Q, Li N, et al., 2024. Enabling mobile AI agent in 6G era: architecture and key technologies. IEEE Netw, 38(5):66-75.

[7]Chen ZR, Zhang ZY, Yang ZH, 2024. Big AI models for 6G wireless networks: opportunities, challenges, and research directions. IEEE Wirel Commun, 31(5):164-172.

[8]Deng ZH, Guo YJ, Han CZ, et al., 2025. AI agents under threat: a survey of key security challenges and future pathways. ACM Comput Surv, 57(7):182.

[9]Du HY, Liu GY, Lin YJ, et al., 2024. Mixture of experts for intelligent networks: a large language model-enabled approach. Proc Int Wireless Communications and Mobile Computing Conf, p.531-536.

[10]Du HY, Zhang RC, Niyato D, et al., 2025. Reinforcement learning with LLMs interaction for distributed diffusion model services.

[11]Dzeparoska K, Lin JY, Tizghadam A, et al., 2023. LLM-based policy generation for intent-based management of applications. Proc 19^th Int Conf on Network and Service Management, p.1-7.

[12]Hou XY, Zhao YJ, Wang SN, et al., 2025. Model context protocol (MCP): landscape, security threats, and future research directions.

[13]Huang SJ, Sun CN, Wang RQ, et al., 2025. Toward adaptive and coordinated transportation systems: a multi-personality multi-agent meta-reinforcement learning framework. IEEE Trans Intell Transp Syst, 26(8):12148-12161.

[14]International Telecommunication Union (ITU), 2023. M. 2160: Framework and Overall Objectives of the Future Development of IMT for 2030 and Beyond. ITU, Geneva.

[15]Jiang FB, Peng YB, Dong L, et al., 2024. Large language model enhanced multi-agent systems for 6G communications. IEEE Wirel Commun, 31(6):48-55.

[16]Jiang XY, Zheng C, Zhuo Y, et al., 2025. Advancing industrial data augmentation in AIGC era: from foundations to frontier applications. IEEE Trans Instrum Meas, 74:1-22.

[17]Li C, Dong SK, Yang SD, et al., 2024. Multi-agent sparse interaction modeling is an anomaly detection problem. Proc IEEE Int Conf on Acoustics, Speech and Signal Processing, p.5890-5894.

[18]Liu Y, Chen WX, Bai YJ, et al., 2025. Aligning cyber space with physical world: a comprehensive survey on embodied AI. IEEE/ASME Trans Mechatron, early access.

[19]Liu YS, Dong XW, Zio E, et al., 2025a. Event-triggered multiple leaders formation tracking for networked swarm system with resilience to noncooperative nodes. IEEE Trans Cybern, 55(9):4136-4144.

[20]Liu YS, Li WX, Dong XW, et al., 2025b. Resilient formation tracking for networked swarm systems under malicious data deception attacks. Int J Robust Nonl Contr, 35(6):2043-2052.

[21]Lu J, Osorio C, 2024. Link transmission model: a formulation with enhanced compute time for large-scale network optimization. Transp Res Part B Methodol, 185:102971.

[22]Mahmoud H, Elbadawy HM, Ismail T, et al., 2024. A comprehensive review of generative AI applications in 6G. Proc 6^th Novel Intelligent and Leading Emerging Sciences Conf, p.593-596.

[23]Ray PP, 2025. A review on agent-to-agent protocol: concept, state-of-the-art, challenges and future directions.

[24]Shahid A, Kliks A, Al-Tahmeesschi A, et al., 2025. Large-scale AI in telecom: charting the roadmap for innovation, scalability, and enhanced digital experiences.

[25]Silva C, Barraca JP, Aguiar R, 2021. eSIM suitability for 5G and B5G enabled IoT verticals. Proc 8^th Int Conf on Future Internet of Things and Cloud, p.210-216.

[26]Sun G, Xu Z, Yu HF, et al., 2021. Dynamic network function provisioning to enable network in box for industrial applications. IEEE Trans Ind Inform, 17(10):7155-7164.

[27]Wang L, Ma C, Feng XY, et al., 2024. A survey on large language model based autonomous agents. Front Comput Sci, 18(6):186345.

[28]Wang XQ, Zhu FH, Yang ZH, et al., 2025. Bridging physical and digital worlds: embodied large AI for future wireless systems.

[29]Wang YD, Chen K, Tan HS, et al., 2023. Tabi: an efficient multi-level inference system for large language models. Proc 18^th European Conf on Computer Systems, p.233-248.

[30]Xu MR, Du HY, Niyato D, et al., 2024a. Unleashing the power of edge-cloud generative AI in mobile networks: a survey of AIGC services. IEEE Commun Surv Tut, 26(2):1127-1170.

[31]Xu MR, Niyato D, Kang JW, et al., 2024b. When large language model agents meet 6G networks: perception, grounding, and alignment. IEEE Wirel Commun, 31(6):63-71.

[32]Xu YF, Chen YN, Zhang XM, et al., 2023. CloudeVal-YAML: a practical benchmark for cloud configuration generation.

[33]Yang A, Li AF, Yang BS, et al., 2025. Qwen3 technical report.

[34]Yu HF, Cui XQ, Zhang H, et al., 2025. fMoE: fine-grained expert offloading for large mixture-of-experts serving. https://arxiv.org/html/2502.05370v1

[35]Zhang MJ, Shen XM, Cao JN, et al., 2025. EdgeShard: efficient LLM inference via collaborative edge computing. IEEE Int Things J, 12(10):13119-13131.

[36]Zhou H, Hu CM, Yuan Y, et al., 2025. Large language model (LLM) for telecommunications: a comprehensive survey on principles, key techniques, and opportunities. IEEE Commun Surv Tut, 27(3):1955-2005.

[37]Zhu FH, Wang XQ, Li XY, et al., 2025. Wireless large AI model: shaping the AI-native future of 6G and beyond.