| 229 | 0 | 30 |
| 下载次数 | 被引频次 | 阅读次数 |
探讨了自智网络在通信业界的普及与演进，并着重分析了两种推动自智网络向高阶发展的关键驱动力：大模型（LLM）和数字孪生技术。大模型在通信网络中的应用展现了其不可替代的优势，并且持续演进的新技术如检索增强生成（RAG）正逐步提升其适配通信网络场景的能力。在实际应用中，大小模型的结合以及多AI智能体的协同工作，为处理复杂场景分析和执行任务提供了有效手段。数字孪生技术则为高阶自智网络的发展提供了控制风险的重要工具，其深化应用需关注的技术焦点包括特征网络模型的精准构建、高效的数据管理技术以及分布式数字孪生系统的完善。大模型和数字孪生技术的融合，不仅有助于网络深入理解用户意图，实现自主决策和执行任务，更为自智网络迈向高阶阶段提供了坚实的技术支撑。
Abstract：The popularity and evolution of autonomous network in the communication industry are introduced, focusing on the analysis of large language model(LLM) and digital twins as key driving forces for advancing autonomous network to a higher level. The application of LLM in communication networks demonstrates their irreplaceable advantages, and evolving technologies such as retrieval-augmented generation(RAG) are gradually enhancing their adaptability to communication network scenarios. In practical applications, the combination of large and small models, along with the collaborative work of multiple AI Agents, provides effective means for handling complex scenario analysis and task execution. Digital twin technology offers an important tool for risk control in the development of high-level autonomous networks, with technical focuses including the precise construction of feature network models, efficient data management techniques, and the refinement of distributed digital twin systems. The integration of LLM and digital twin technology not only helps networks deeply understand user intentions, make autonomous decisions, and execute tasks, but also provides solid technical support for autonomous network to move towards a higher stage.
[1] LIU G Y, LI N, DENG J, et al. The SOLIDS 6G mobile network architecture:driving forces, features, and functional topology[J].Engineering, 2022, 8:42-59. DOI:10.1016/j.eng.2021.07.013
[2]中国移动.中国移动自智网络白皮书(2023)[R]. 2023
[3] TM Forum. Autonomous networks:empowering digital transformation[R].2023
[4] OpenAI. Chat GPT plugins[EB/OL].[2024-06-15]. https://openai.com/index/chatgpt-plugins/
[5] LEWIS P, PEREZ E, PIKTUS A, et al. Retrieval-augmented generation for knowledge-intensive NLP tasks[EB/OL].[2024-06-15]. http://arxiv.org/abs/2005.11401
[6] BULB. Retrieval-augmented generation(RAG):from theory to langchain implementation[EB/OL].[2024-06-15]. https://www.bulbapp. io/p/46eb00e7-c08e-447c-9bc6-283edae5e889/retrievalaugmented-generation-rag-from-theory-to-langchainimplementation
[7] MASTERMAN T, BESEN S, SAWTELL M, et al. The landscape of emerging AI agent architectures for reasoning, planning, and tool calling:a survey[EB/OL].[2024-06-15]. https://arxiv. org/abs/2404.11584
[8] YAKLAF S K A, TARMISSI K S, SHASHOA N A A. 6G mobile communications systems:requirements, specifications, challenges,applications, and technologies[C]//Proceedings of IEEE 1st International Maghreb Meeting of the Conference on Sciences and Techniques of Automatic Control and Computer Engineering MI-STA. IEEE, 2021:679-683.DOI:10.1109/MISTA52233.2021.9464470
[9] SHU M, SUN W F, ZHANG J, et al. Digitaltwin-enabled 6G network autonomy and generative intelligence:architecture,technologies and applications[J]. Digital twin, 2022, 2:16. DOI:10.12688/digitaltwin.17720.1
[10] WANG D, SUN T, SUN X, et al. Study on digital twins in network lifecycle management[J]. Telecommunication science. 2022; 38(4):138-145. DOI:10.11959/j.issn.1000-0801.2022063
[11] ZHANG M, TAO F, HUANG B Q, et al.Digital twin data:methods and key technologies[J]. Digital twin, 2022, 1:2.DOI:10.12688/digitaltwin.17467.1
[12] YANG B, WANG S L, LI S, et al. Digital thread-driven proactive and reactive service composition for cloud manufacturing[J]. IEEE transactions on industrial informatics, 2023, 19(3):2952-2962. DOI:10.1109/TII.2022.3171338
[13]Gartner.Top trends in data and analytics for 2021:data fabric is the foundation[EB/OL].(2021-02-16)[2024-06-15].https://www.gartner.com/en/documents/3996983
[14]中国移动.支撑通信运维大模型及自智网络的通信系统端到端网络数字孪生系统[R].2024
[15]TM Forum.Digital twin for network operations-phase II[EB/OL].(2021-02-16)[2024-06-15].https://www.tmforum.org/catalysts/projects/C23.0.530/
基本信息:
DOI：
中图分类号:TN929.5
引用信息:
[1]孙方平,钱铮铁.高阶自智网络关键技术及应用[J].中兴通讯技术,2024,30(04):77-82.
基金信息: