Intelligent optical network architecture and key technologies driven by data and model collaboration

AbstractThe network scale upgrade and ultra-large connections, ultra-high bandwidth, ultra-low latency The continuous deepening of optical transmission network applications has put forward higher requirements for the utilization of optical transmission network resources and differentiated network services, making the network form and configuration methods driven by traditional models facing challenges. Based on the idea of ​​collaborative driving of data and models, a "3-layer 3-loop" architecture and an intelligent optical network technology solution with "3 functions" characteristics are proposed, and research on intelligent implementation technology is carried out, and all intelligent transmission network platforms are developed and designed. The performance of the proposed algorithm was tested, and it was verified that the transmission performance of intelligent optical networks driven by data and model collaboration has been effectively improved, providing theoretical and technical support for realizing network intelligence.

KeywordsIntelligent optical network; data and model collaborative drive; physical layer status awareness ; Network layer resource linkage; Business layer slice customization

0 Preface

The scale of communication network deployment is constantly upgrading, and the traffic demand of broadband networks is increasing day by day, especially the emergence of 4K/8K video, virtual reality (VR) games, Internet of Things, autonomous driving and other related intelligent terminals and new services. New and differentiated requirements have been put forward for network transmission performance in terms of bandwidth, delay, reliability and other aspects. As an optical transmission network that carries services, although its performance in terms of transmission capacity and transmission quality has been widely recognized, it still faces many challenges such as insufficient transmission resources to meet the needs of future emerging services. Therefore, how to achieve efficient integrated utilization of resources and provide differentiated and customized services has become an urgent problem to be solved in the development of transmission networks.

In response to the above problems, the conventional idea is usually to improve the "hard power" of the network. On the one hand, the transmission capacity of the network is increased through multi-mode, multi-core and other methods. On the other hand, the traditional fixed grid is transformed into a flexibly adjustable grid, and the rigid pipe is transformed into a flexible pipe to improve the utilization of optical network resources. However, the transmission capacity and efficiency of broadband networks established under traditional mathematical models and physical laws such as Shannon's Theorem and Moore's Law have tended to "

Guess you like

Origin blog.csdn.net/weixin_57147647/article/details/134901244