期刊文献+

6G通算融合网络架构 被引量:4

6G Network Architecture for Communication and Computing Integration
下载PDF
导出
摘要 边缘计算在物理位置上部署在通信网络内,但是在逻辑上计算与通信拥有彼此独立的管控体系,协同是非实时的,在应对超低时延业务场景下用户信道状态、用户移动性等的动态变化时,存在技术挑战。因此,6G网络架构在设计之初,就需要原生的支持通信与计算的深度融合机制,基于此,提出了一种在无线网络内通信资源与分布式计算资源实时协同的计算面方案,来实现6G网络业务适配能力上的突破,提升综合资源能效。在终端和基站联合模型拆分推理场景下,通过仿真对比了控制面融合和管理面融合技术方案,验证了控制面融合方案能更好地应对终端空口连接带宽发生的变化,大幅降低业务时延抖动。 Edge computing can be deployed in communications network physically.However,computing and communication have independent management and control systems.As a result,it is hard for the non-real-time collaboration scheme to solve the challenges caused by dynamic changes of user channel status,user mobility in the scenario with ultra-low-latency requirement.Therefore,a mechanism of native communication and computing integration should be supported at the beginning of the design of the 6G network architecture.This paper proposes a computing plane solution that coordinates communication resources and distributed computing resources of wireless network in real-time.It can improve the service adaptation capability and efficiency of heterogeneous resources of the 6G network.In the scenario of model-splitting inference between terminal and base station with the changed bandwidth,the simulation results shows that the control plane convergence solution can achieve better delay and jitter compared with management plane convergence solution.
作者 彭程晖 邓娟 吴建军 刘哲 李芳芳 刘光毅 孙韶辉 杨旸 张宏纲 李荣鹏 PENG Chenghui;DENG Juan;WU Jianjun;LIU Zhe;LI Fangfang;LIU Guangyi;SUN Shaohui;YANG Yang;ZHANG Honggang;LI Rongpeng(Wireless Technology Lab,Huawei Technologies Co.,Ltd.,Shanghai,201206,China;Future Research Lab,China Mobile Research Institute,Beijing 100053,China;CITIC Mobile Communication Technology Co.,Ltd.,Beijing 100083,China;School of Information Science and Technology,Shanghai University of Science and Technology,Shanghai 201210,China;College of Electronic and Information Engineering,Zhejiang University,Hangzhou 310027,China)
出处 《无线电通信技术》 2022年第4期583-591,共9页 Radio Communications Technology
基金 国家重点研发计划(2020YFB1806800) 国家自然科学基金企业创新发展联合基金(U21B2002)。
关键词 网络AI 通算融合 计算面 无线通信网络 network AI communication and computing integration computing plane wireless communication network
  • 相关文献

参考文献5

二级参考文献28

共引文献133

同被引文献40

引证文献4

二级引证文献14

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部