Digital Twin in 6G: Embracing Comprehensive Network Intelligence

While 5G has opened the era of the Internet of everything (IoE), the 5G cellular systems do not have the pervasive intelligence of devices considered in their original design. In the era of 6G, due to the advance of artificial intelligence (AI), not only have the connected end hosts changed from electronic devices to autonomous AI systems, but also the network equipment (e.g., switches, routers) has the capability of computing and AI. With the three key components of communications, that is, information source, destination, and transmission channel, making fundamental changes to be AI-enabled, the 6G network is destined to be a new communication system of comprehensive AI. In this article, we investigate a practical network framework supported by digital twin (DT) to enable comprehensive AI and facilitate comprehensive network intelligence for 6G. We first introduce the basic concepts of DT and propose a new communication model for 6G. Based on this, a DT-empowered architecture that can fully use the pervasive AI in 6G is proposed with the enabled technologies and promising applications. As a case study, we investigate task offloading in vehicular DT networks, and design an efficient task scheduling mechanism. Finally, open issues are discussed to envision the challenges and opportunities of DT in the future 6G network.