面向AI Agent互联网的计算型智能宽带网络网关架构及应用探讨

翁思俊; 颜宇; 杨新国; 宋健; 姜文颖; 程伟强

doi:10.11959/j.issn.1000-0801.DXKX260133

Chinese

您当前的位置：

首页 >

文章列表页 >

面向AI Agent互联网的计算型智能宽带网络网关架构及应用探讨

更新时间：2026-05-07

- 面向AI Agent互联网的计算型智能宽带网络网关架构及应用探讨
- Discussion on the Architecture and Application of Computing Intelligent Broadband Network Gateway for the AI Agent Internet
- 电信科学 2026年
- 作者机构：
  
  中国移动通信有限公司研究院，北京，100053
- 作者简介：
  
  [ "翁思俊（1988），女，中国移动研究院基础网络技术研究所研究员，主要从事IP承载网及智算IP网络相关领域技术研究、标准制定和网络推进；" ]
  [ "颜宇（1978），女，中国移动研究院基础网络技术研究所项目经理，主要从事IP城域网、骨干网领域相关技术研究、标准制定及现网推进工作；" ]
  [ "杨新国（1981），男，中国移动研究院基础网络技术研究所项目经理，主要从事家庭宽带业务及IP城域网领域相关技术研究、标准制定及现网推进工作；" ]
  [ "宋健，（1983），男，中国移动研究院基础网络所项目经理，主要从事智算 IP 网络新架构新业务的洞察分析和技术规划等工作；" ]
  [ "姜文颖（1977-），女，中国移动研究院基础网络技术研究所承载网室技术经理，主要从事下一代互联网、IPv6+基础协议、AI广域网、超高速以太网等；" ]
  [ "程伟强（1980），男，中国移动研究院基础网络网络技术研究所副所长，教授级高级工程师，长期从事网络领域的技术研究和标准制定工作。在ITU-T、IETF、ONF等国际标准组织中主导发布20余项国际标准，任BBF SPAC委员会副主席、IETF SRv6头压缩设计组主席，曾任ONF电信级SDN工作组主席。" ]
- 基金信息：
- DOI：10.11959/j.issn.1000-0801.DXKX260133
  中图分类号： TP393
- 收稿：2026-02-28，
  
  修回：2026-03-19，
  
  录用：2026-04-16，
- 稿件说明：
移动端阅览
翁思俊, 颜宇, 杨新国, 等. 面向AI Agent互联网的计算型智能宽带网络网关架构及应用探讨[J/OL]. 电信科学, 2026.

WENG Sijun, YAN yu, YANG Xinguo, et al. Discussion on the Architecture and Application of Computing Intelligent Broadband Network Gateway for the AI Agent Internet[J/OL]. Telecommunications Science, 2026.
翁思俊, 颜宇, 杨新国, 等. 面向AI Agent互联网的计算型智能宽带网络网关架构及应用探讨[J/OL]. 电信科学, 2026. DOI： 10.11959/j.issn.1000-0801.DXKX260133.

WENG Sijun, YAN yu, YANG Xinguo, et al. Discussion on the Architecture and Application of Computing Intelligent Broadband Network Gateway for the AI Agent Internet[J/OL]. Telecommunications Science, 2026. DOI： 10.11959/j.issn.1000-0801.DXKX260133.

摘要

生成式AI与大语言模型推动互联网向AI Agent互联网演进。宽带网络网关（BNG）作为宽带用户接入网络的关键设备，难以适配海量异构AI Agent超大规模管理与边缘算力的新需求。智能宽带网络网关（iBNG）是传统BNG的智能化升级形态，但其仍面临系列挑战。作为前瞻性理论研究，本文面向AI Agent互联网，提出计算型iBNG创新架构，系统阐述其技术架构与关键技术体系，并结合智慧家庭、无人商业、小型工业三大典型场景探讨应用路径。该架构深度融合“联接-计算-智能”能力，通过内生算力与智能管控一体化设计，为AI Agent互联网提供系统性接入网关解决方案。

Abstract

The evolution of the Internet toward the AI Agent Internet was driven by generative AI and large language models. As a key device for broadband user access to the network

the broadband network gateway (BNG) was difficult to adapt to the new requirements of ultra-large-scale management and edge computing power posed by massive heterogeneous AI Agents. The intelligent broadband network gateway (iBNG) was proposed as an intelligent upgrade of the traditional BNG

yet a series of challenges were still faced. As a forward-looking theoretical exploration

this paper proposes an innovative computing iBNG architecture for the AI Agent Internet

systematically elaborates its technical architecture and key technology system

and discusses its application paths in three typical scenarios: smart home

unmanned commerce

and small-scale industry. By deeply integrating the capabilities of ‘connection

computing

and intelligence

’ this architecture provides a systematic access gateway solution for the AI Agent Internet through the integrated design of endogenous computing power and intelligent management and control.

关键词

Keywords

references

ZHANG R , LIU G , LIU Y , et al . Toward Edge General Intelligence with Agentic AI and Agentification: Concepts, Technologies, and Future Directions [J ] . 2025 . DOI: 10.1109/COMST.2026.3651702 http://dx.doi.org/10.1109/COMST.2026.3651702 .

XU M R , NIYATO D , KANG J W , et al . When large language model agents meet 6G networks: perception, grounding, and alignment [J ] . IEEE Wireless Communications , 2024 , 31 ( 6 ): 63 - 71 .

LI B R , LIU T E , WANG W L , et al . Agent-as-a-service: an AI-native edge computing framework for 6G networks [J ] . IEEE Network , 2025 , 39 ( 2 ): 44 - 51 .

XIAO Y , SHI G , ZHANG P . Towards agentic AI networking in 6G: a generative foundation model-as-agent approach [EB/OL ] . ( 2025-03-01 )[ 2025-02-11 ] . arXiv: 2503.15764 .

HOU X , ZHAO Y , WANG S , et al . Model context protocol (MCP): landscape, security threats, and future research directions [EB/OL ] . ( 2025-03-01 )[ 2025-02-11 ] . arXiv: 2503.23278 .

TUPE V , THUBE S . Demonstrating multi-agent collaboration via agent-to-agent and model context protocols: an IT incident response case study [C ] // Proceedings of 2025 IEEE International Conference on Service-Oriented System Engineering . Piscataway : IEEE Press , 2025 : 1 - 5 .

ZHOU Z , CHEN X , LI E , et al . Edge intelligence: paving the last mile of artificial intelligence with edge computing [J ] . Proceedings of the IEEE , 107 ( 8 ): 1738 - 1762 .

DENG S , ZHAO H , FANG W , et al . Edge intelligence: the confluence of edge computing and artificial intelligence [J ] . IEEE Internet of Things Journal , 2020 , 7 ( 8 ): 7457 - 7469 .

FRIHA O , FERRAG M A , KANTARCI B , et al . LLM-based edge intelligence: a comprehensive survey on architectures, applications, security and trustworthiness [J ] . IEEE Open Journal of the Communications Society , 2024 , 5 : 5799 - 5856 .

LIN J , WANG H , CHEN Y , et al . SLED: A Speculative LLM Decoding Framework for Efficient Edge Serving [C ] // SEC 2025: Tenth ACM/IEEE Symposium on Edge Computing . Piscataway : IEEE Press , 2025 : 1 - 12 .

王睿 , 张留洋 , 高志涌 , 姜彤雲 . 面向边缘智能的大模型研究进展 [J ] . 计算机研究与发展 , 2025 , 62 ( 10 ): 2565 - 2582 .

KHAN L U , YAQOOB I , TRAN N H , et al . Edge computing enabled smart cities: a comprehensive survey [J ] . IEEE Internet of Things Journal , 2020 , 7 ( 10 ): 10200 - 10232 .

PORAMBAGE P , OKWUIBE J , LIYANAGE M , et al . Survey on multi-access edge computing for internet of things realization [J ] . IEEE Communications Surveys & Tutorials , 2018 , 20 ( 4 ): 2961 - 2991 .

LI M , ZHANG Y , WANG Q , et al . Power internet of things networked group intelligent sensor networking [J ] . Journal of Physics: Conference Series , 2021 , 1744 ( 1 ): 012345 .

KATO N , FADLULLAH Z M , TANG F , et al . The deep learning vision for heterogeneous network traffic control: proposal, challenges, and future perspective [J ] . IEEE Wireless Communications , 2017 , 24 ( 3 ): 146 - 153 .

工业和信息化部。智能化宽带网络网关（iBNG）管控系统技术要求：2024‑0500T‑YD [S ] . 北京：中国标准出版社， 2025 .

工业和信息化部。智能化宽带网络网关（iBNG）架构及总体技术要求：2023‑11087176‑YD [S ] . 北京：中国标准出版社， 2024

BROADBAND FORUM . DBNG for Wired Access (WT‑487) [S ] . Brussels ： Broadband Forum , 2024 .

BROADBAND FORUM . Broadband Forum. Multi-Service Disaggregated BNG with CUPS: Reference Architecture, Deployment Models, Interface, and Protocol Specification (TR-459 Issue 3) [S ] . 2025 .

BROADBAND FORUM . Multi-Service Disaggregated BNG with CUPS: Integrated Carrier Grade NAT Function (TR-459.2 Issue 2) [S ] . 2025 .

KATENDE R . Rethinking data-efficient artificial intelligence for low-resource settings [J ] . Machine Learning with Applications , 2026 , 23 : 100796 .

HUANG H H . Democratizing LLM efficiency: from hyperscale optimizations to universal deployability [J/OL ] . ArXiv , 2025 , abs/2511.20662. https://api.semanticscholar.org/CorpusID:283261746 https://api.semanticscholar.org/CorpusID:283261746 .

SAPKOTA R , ROUMELIOTIS K I , KARKEE M . AI Agents vs. Agentic AI: A Conceptual taxonomy, applications and challenges [J ] . Information Fusion , 2026 , 126 ( B ): 103599 .

刘阳，柏永杰，林倞。面向人机物高效融合与协作的具身智能技术体系 [J ] . 机器人， 2025 , 47 ( 4 ): 560 - 580 .

SimilarWeb . SimilarWeb: DeepSeek 周访问量激增 614 %，峰值访问量达4900万次[EB/OL ] . 2025 . https://www.similarweb.com/zh-cn/blog/deepseek-traffic-surge/ https://www.similarweb.com/zh-cn/blog/deepseek-traffic-surge/ .

中国第一汽车集团有限公司。基于 5G-V2X 的 L4 级智能商用车车路协同系统 [EB/OL ] . 2024 . http://wap.sasac.gov.cn/n4470048/n29955503/n30143035/n30143047/n30143160/c30163453.html http://wap.sasac.gov.cn/n4470048/n29955503/n30143035/n30143047/n30143160/c30163453.html .

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

暂无数据