Intelligence-Endogenous Networks: Innovative Network Paradigm for 6G

Abstract

Feature applications envisioned in the 6G era, represented by holographic communication, full-sensing communication, intelligent transportation, and intelligent manufacturing in terms of virtual and physical integration, immersion, contextualization, and personalization, have placed high demands on the intelligentization of 6G networks. The current network operating mechanism based on rule-based algorithms is limited by rigid preset rules. It is difficult to adapt dynamically to constantly changing user needs and network environments. In order to enable the network to effectively accumulate and independently utilize the operating experience and realize the independent iterative evolution of network operation, management, and control capabilities, this article proposes the concept of the intelligence-endogenous network (IEN). The IEN aims to introduce knowledge graph and artificial intelligence technologies into networks to characterize, construct, learn, and apply the network's multidimensional subjective and objective knowledge. Based on the learned knowledge, an IEN can realize stereo perception, decision deduction, and dynamic adjustment, so the network can change as it needs for whatever new services we want. The article introduces the basic concepts, goals, and meanings of the IEN, together with its key principles and technologies, and illustrates how it innovatively responds to the current network's multifaceted challenges with an intuitive use case.