摘要
未来6G网络将内生支持通信和AI一体化服务,赋能丰富多彩的新业务,支撑社会高效可持续发展。为此,借鉴了IT行业AI Agent的应用范式,基于电信应用场景创新地提出了6G AI Agent技术框架的三大设计理念,包括多模型融合、定制化Agent和插件式环境交互,并基于该理念构建了6G AI Agent技术框架。通过环境交互层、Agent引擎层、模型调度层、模型基座层交互协同,实现了自主环境感知、自主任务生成和自主执行任务的能力。此外,以移动网络的智能感知任务为例,探索了AI Agent的使用场景及价值,为AI新技术在电信领域发展提供了新的思路和技术支撑。
In the future,6G network will inherently support integrated communications and AI services,empower a variety of new services,and support efficient and sustainable development of society.So we drew on the AI Agent application paradigm in the IT industry and innovatively proposed three design concepts for the 6G AI Agent technology framework based on telecommunications application scenarios,including multi-model fusion,customized Agent and plug-in environment interaction,and based on this concept Constructed the 6G AI Agent technology framework.Through the interaction and collaboration of the environment interaction layer,Agent engine layer,model scheduling layer,and model base layer,the capabilities of autonomous environment perception,autonomous task generation,and autonomous task execution are realized.In addition,taking the intelligent sensing task of mobile networks as an example,the usage scenarios and value of AI Agent are explored,providing new ideas and technical support for the development of new AI technologies in the telecommunications field.
作者
陈新宇
王卫斌
陆光辉
CHEN Xinyu;WANG Weibin;LU Guanghui(ZTE Corporation,Nanjing 210012,China)
出处
《移动通信》
2024年第7期28-32,共5页
Mobile Communications
基金
国家重点研发计划项目“多模态网络与通信”(2022YFB2901602)。