Knowledge-Defined Networking

To achieve the “intelligent endogeneity” of 6G, it is necessary to fully integrate AI into the network from the outset of the design of the 6G network so that its “intelligence” characteristics are naturally reflected at all levels of the network. To achieve this goal, it is imperative to introduce a structure that performs “brain” functions into the 6G network. This structure would be able to perceive changes inside and outside the network at any time and direct the network actions based on these changes. More importantly, it would be capable of continuously learning new knowledge, updating its own knowledge system to better cope with unknown changes and make forward-looking judgments. One of the mainstream directions for designing, enriching, and improving the functions of the “brain” of the 6G network is based on Knowledge-Defined Networking (KDN). This chapter will introduce KDN technology, starting with the proposal and basic concepts of KDN; then it will discuss the architecture, process and knowledge acquisition of KDN; finally, it will introduce the application scenarios of KDN.