As the importance of Digital Twin Network (DTN) has become a common sense in the study of 5G, 6G, and future network, the means to implement DTN and the performance challenges in DTN start to be discussed. However, the reproduction compression of DTN, i.e. the design and compression requirements of n:1 reproduction, is a technical gap. Additionally, there is no corresponding technical solution to the time delay of network intelligence calling DTN system. A Tiered Native Digital Twin for Telco (TNDT4Telco) is proposed for future communication systems in this paper, based on the multi-level distributed design philosophy. TNDT4Telco resolves two key issues in DTN design, i.e., n:1 reproduction against huge quantity of raw data, and DTN wake-up time. Numerical evaluation shows that TNDT4Telco can provide 36% total DTN wake-up time saving, and around 10 times gain with different number of parallel DTN tasks in DTN task aggregation performance.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

A Tiered Native Digital Twin for Telco Networks

  • Shoufeng Wang,
  • Huamin Chen,
  • Jianchao Guo,
  • Ye Ouyang,
  • Fan Li,
  • Xuan Chen,
  • Sen Bian,
  • Zhongke Zhang,
  • Lianhua Zhang,
  • Yun Li

摘要

As the importance of Digital Twin Network (DTN) has become a common sense in the study of 5G, 6G, and future network, the means to implement DTN and the performance challenges in DTN start to be discussed. However, the reproduction compression of DTN, i.e. the design and compression requirements of n:1 reproduction, is a technical gap. Additionally, there is no corresponding technical solution to the time delay of network intelligence calling DTN system. A Tiered Native Digital Twin for Telco (TNDT4Telco) is proposed for future communication systems in this paper, based on the multi-level distributed design philosophy. TNDT4Telco resolves two key issues in DTN design, i.e., n:1 reproduction against huge quantity of raw data, and DTN wake-up time. Numerical evaluation shows that TNDT4Telco can provide 36% total DTN wake-up time saving, and around 10 times gain with different number of parallel DTN tasks in DTN task aggregation performance.